tag:blogger.com,1999:blog-111685552024-03-16T14:52:57.233-04:00The Multiverse According to BenBenhttp://www.blogger.com/profile/12743597120529571571noreply@blogger.comBlogger239125tag:blogger.com,1999:blog-11168555.post-91618146676642268962022-07-16T11:49:00.001-04:002022-07-16T11:49:25.923-04:00The Plight of the Humans<p> I haven't written song lyrics for quite some years but was recently inspired to update the lyrics of a song I wrote back in the late 90s or early aughts -- "Plight of the Humans" aka "I see it"<br /><br />Some of the newer portions of the lyrics are morphed from lines in the <a href=" https://www.amazon.com/Plight-Humans-Jesucopter-Crudslinger/dp/150554758X" target="_blank">most insane prose-poem</a> any of my subpersonalities ever came up with,<br /><br />A pretty crude, off-the-cuff <a href="https://www.youtube.com/watch?v=VRcr9YCH6vw">solo version of the song is here</a>, ... in time there may come a version with <a href="http://jamgalaxy.com" target="_blank">Jam Galaxy</a> Band (see our <a href="https://media.namm.org/nso22/JamGalaxyBand.mp4" target="_blank">gig at NAMM</a> a couple months ago here, -- this was a short set so all songs were reduced to 6 min or so; soon there will be video/audio from our follow-up gig at DROM in New York where things were improvised and outprovised at much greater length... and another gig in the metaverse, and at <a href="https://www.rarebloom.io/" target="_blank">Rare Bloom</a> Cardano communityevent ... )<br /><br />The phrase "circus of the empty eye" is cribbed from <a href="https://www.onlinereadfreebooks.com/en/The-Poems-of-Octavio-Paz-255460/25" target="_blank">Octavio Paz</a>; the phrase "shadow of a dream" from <a href="https://www.kiplingsociety.co.uk/poem/poems_mcandrew.htm" target="_blank">Rudyard Kipling</a>. <br /><br /><br />***<br /><b>THE PLIGHT OF THE HUMANS</b><br />***</p><p><br /><span style="background-color: white; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; white-space: pre-wrap;"><i>
I see it in the delicate Martian gleam of the distant mountains</i></span></p><p><i style="color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; white-space: pre-wrap;">I see it in the geometry of the car crash in my eyes</i></p><p><i style="color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; white-space: pre-wrap;">I see it in the sunrise bleeding love and death and magic through the frights that polarize</i></p><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it in the mottled fire in the soul of the alligator</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it in the wet dream of the four trillionth femto-memristor</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it in the glory of human flesh as it loses its measure and its mind</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>What else to find</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><br /></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><br /></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it in the protocol of awakening and dissolving</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>In the calorimeter that measures the expansion and contraction of our sighs</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it in the love that weaves individuation and self-transcendence</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Into black and white holes whirling singularities through the sick bowels of the night</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it in the howling hound guarding the exit of humanity</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it in the passionate, deranged mammal topology of her smile</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Of her smile sweet smile</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it in the irresistible force too strong for specification</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it in the Drumcloud Dionysus versus the Crucified</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it in the glimmering needles weaving the cosmic background fabric of the her shrieks converged in time</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it in the furious abandon of utter metaphysical nudity gleaming crystalline in the double-sun sky</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it in the heart of the heartless, the soul of the soulless, the mind of the the mindless</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>The beauty in the center of the madness of the center of the circus of the empty eye
<br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>The circus of the empty eye</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it, I see it</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it, I see it</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Everything</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Is just the shadow of a dream</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Everything</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Is just the shadow of a dream</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Everything we do and feel</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>And hear and think and see</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Is just the shadow of the shadow of a dream</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>To me</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Everything we do and feel</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>And hear and think and see</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Is just the shadow of the shadow of a dream</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>To you and me</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it in the screams of a billion brachycephalic ice floes</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>as they stretch like magic-mushroom membranes across the Multiversal Eye</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it in the Law of Madness delivering discipline in the dung-heap,</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>In the ghettoes, in the prison, in young love and ancient blight</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it in the truth that follows from the fallacy that leads the army of paraconsistent thought-sparks</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>All venting chaos through their memes</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>And stenting veins in the necrotic imagination</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Of the Universal Mind</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it, wrought with fever, sporked with pain and funked with glee</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>In the ecstatic torture of being me, and knowing I'll never be me</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it in the robot mouthing words of necromantic need and spastic subtlety</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>While my fingers stagger across the keys</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Of black and white and perjury</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it, I see it</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>I see it, I see it</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Everything</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Is just the shadow of a dream</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Everything</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Is just the shadow of a dream</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Everything we do and feel</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>And hear and think and see</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Is just the shadow of the shadow of a dream</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>To me</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Everything we do and feel</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>And hear and think and see</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Is just the shadow of the shadow of a dream</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>To you and me</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>All the love and hate that gets</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Unfolded and unfurled</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Is just the shadow of a dream</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Here in this simulated world</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>All these words, ideas, sounds</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Emotions and inclines</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Are just the shadow of a dream</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>In an imaginary mind</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Every fantasy that weaves</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Its wonder through your mind</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Is just a shadow of a shadow</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>In illusory spacetime</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Everything we do and feel</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>And hear and think and see</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>Is just the shadow of the shadow of a dream</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i><br /></i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><i>To you and me</i></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><br /></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><br /></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><br /></div><div style="background: rgb(255, 255, 255); border: 0px; color: #0d0d0d; font-family: Roboto, Noto, sans-serif; font-size: 15px; margin: 0px; padding: 0px; white-space: pre-wrap;"><br /></div>Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com19tag:blogger.com,1999:blog-11168555.post-18836875165113693762020-11-25T18:50:00.004-05:002020-11-27T02:15:40.241-05:00Eternal Childhood of the Post-Singularity Mind<p style="text-align: left;"><span style="font-family: arial;">It has occurred to me many times, as I played with my various offspring (4 kids and one grandchild so far, one more little daughter due February!), that in many respects our post-Singularity selves are going to closely resemble children today.</span></p><p><span style="font-family: arial; white-space: pre-wrap;">When children play, they’re not just randomly experimenting and exploring and enjoying themselves -- they are doing that, but they’re also whimsically rehearsing and practicing stuff that they (consciously and unconsciously) know they’re potentially going to be doing “for real” once they grow up. They’re practicing, and they’re role-playing roles that they at least half-know they could really get called on to play in the future. But they’re doing this in a way that’s often more fun than the real thing -- because it doesn’t come with all the constraints and risks that the real thing does. </span></p><p><span style="font-family: arial; white-space: pre-wrap;">So when a kid plays “house”, or plays “fighting dragons” -- on the one hand they know one day they may be managing a real house or fighting some actual dangerous thing. On the other hand, they also know in the back of their minds how tedious or scary these real activities might be -- and that their make-believe versions are joyous mockeries of the real thing.</span></p><p><span style="font-family: arial; white-space: pre-wrap;">Implicit in childhood play is a half-knowledge of what one is going to become -- after one gets radically transformed beyond what one is -- and a desire to creatively and lightheartedly explore these future potentials .</span></p><p><span style="font-family: arial; white-space: pre-wrap;">At the same time, as a child, one also knows that one is being cared for by other beings who understand the world much better than one does in important respects -- and have superior power -- and who are setting the context for one’s play activities.</span></p><p><span style="font-family: arial; white-space: pre-wrap;">Now put yourself in the position of a post-Singularity human, who is in a world populated by AGIs with far greater general intelligence and capability, most of whom are basically benevolent toward you. </span></p><p><span style="font-family: arial; white-space: pre-wrap;">In such a situation there are several sorts of opportunities open to you: you can remain a legacy human, you can rapidly upgrade yourself to massively superhuman effective-godhood, or you can gradually upgrade yourself, retaining a sense of self and continuity as you grow and transcend.</span></p><p><span style="font-family: arial; white-space: pre-wrap;">Let’s consider the latter option: So, suppose you are in the process of gradually upgrading your own intelligence and capability. </span></p><p><span style="font-family: arial; white-space: pre-wrap;">Suppose you have chosen to upgrade your intelligence relatively slowly -- say, multiplying it by a factor of 1.5 or 2 each year -- so as to get the full and rich experience of growing and self-transcending (as opposed to the process of increasing one’s intelligence by a factor of 1000 in one second, which in essence would be a kind of death in which the self is suddenly replaced by a radically different entity with a subtle connection to the previous version).</span></p><p><span style="font-family: arial; white-space: pre-wrap;">Then in key respects, your experience will be like that of a human child. </span></p><p><span style="font-family: arial; white-space: pre-wrap;">You’ll be in a world mostly guided by minds far beyond your own -- and bearing some resemblance to what you expect to eventually evolve into. </span></p><p><span style="font-family: arial; white-space: pre-wrap;">And you’ll be highly tempted to play at the activities that you know you’ll be able to partake in more fully once you’ve upgraded a bit more. While at the same time enjoying the activities you can currently partake in with full success.</span></p><p><span style="font-family: arial; white-space: pre-wrap;">By playing at the activities your future self will master, you’ll integrate aspects of these activities into your self, thus smoothing the transition between your current self and your future more-intelligent self.</span></p><p><span style="font-family: arial; white-space: pre-wrap;">Simple example: If you know that a year later you’ll be able to sculpt complex virtual reality worlds rapidly and near-instantly at will, keeping all the different aspects of the world in your mind at once -- then as this ability gradually emerges for you, you’ll likely feel drawn to experiment with world-building even though you’re not that good at it yet. </span></p><p><span style="font-family: arial; white-space: pre-wrap;">Two years of intelligence-evolution away from being able to cognitively interface directly with sensors at the quark level, and grok some of the thoughts of the intra-particulate intelligences resident there? Won’t you be curious what this is like -- and eager to dip now and then into the now-mostly-incomprehensible sensor output -- and creatively improvise on what snippets of meaning one has been able to extract?</span></p><p><span style="font-family: arial; white-space: pre-wrap;">Once humanity finally grows up and moves from the economy of scarcity into the economy of abundance, and we get root access to our own brains and minds and the ability to actively sculpt our own futures -- then we will enter a new era of childhood, one that neither terminates in the relative stasis of an adulthood nor gets repetitive as it would if one remained a typical human eight year old kid forever. </span></p><p><span style="font-family: arial; white-space: pre-wrap;">In this sort of childhood, the moment one seriously approaches the sort of maturity one can understand, one sees that there is a yet higher form of maturity, with respect to which one is just a little child again.</span></p><p><span style="font-family: arial; white-space: pre-wrap;">But wait a minute -- isn’t this our situation anyway? It is, of course. At age 53, I haven’t stopped growing and advancing and changing, and in some ways I’ve personally evolved more in the last 5 years than I did in the 20 years before that. Even in our current human situation, without AGIs or cyborgs, the more we mature our understanding and consciousness, the more clearly we see what the next levels of consciousness are, out there on the horizon.</span></p><p><span style="font-family: arial; white-space: pre-wrap;">But still there are senses in which human development does slow down once adulthood sets in. While the mind can keep advancing, the brain mostly stops developing in the early teens. The last really exciting physical change we experience is puberty -- the changes that occur as one meets old age are generally significantly less inspiring. </span></p><p><span style="font-family: arial; white-space: pre-wrap;">Post-Singularity, our bodies as well as our minds will keep gaining new capabilities, and the computing infrastructure supporting our thinking and feeling activity will keep getting richer and more powerful -- and these improvements will unfold in sync with our advances in cognition and consciousness, just as happens in human childhood.</span></p><p><span style="font-family: arial;">Given the myriad difficulties of human life on Earth right now, and the likelihood of a difficult transition from here to Singularity, it can sometimes be hard to believe we could be so close to such radical positive changes. But this is the nature of nonlinear processes and exponential advance. </span> </p><p><span style="font-family: arial;"><span style="font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">As a kid I loved Arthur C. Clarke’s novel </span><span style="font-style: italic; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Childhood’s End</span><span style="font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">. But in fact, I think the Singularity will bring an acceleration and enhancement of the human childhood experience, rather than its termination.</span></span></p><p><span style="font-family: arial; white-space: pre-wrap;"><b><i>Here’s to the eternal childhood of the post-Singularity Mind!</i></b></span></p><p><br /><br /><br /><br /><br /><br /><br /><br /></p>Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com76tag:blogger.com,1999:blog-11168555.post-18509402148162275312020-09-14T02:54:00.002-04:002020-09-14T03:00:52.176-04:00 Sand Talk, Singularity and Psi<p><br /></p><span id="docs-internal-guid-7860ecdf-7fff-bf0d-6d87-6c8d87183c30"><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">It’s not that often I encounter a newly authored book that I consider a “must read” and avidly recommend to all my friends and colleagues -- but </span><a href="https://www.amazon.com/Sand-Talk-Indigenous-Thinking-World/dp/0062975641" style="text-decoration-line: none;"><span style="background-color: white; color: #1155cc; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; text-decoration-line: underline; text-decoration-skip-ink: none; vertical-align: baseline; white-space: pre-wrap;">Sand Talk</span></a><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;"> by Tyson Yunkaporta genuinely falls into this category. </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">I was introduced to the book by Jim Rutt who<a href="https://jimruttshow.blubrry.net/tyson-yunkaporta-1/"> interviewed the author on his podcast </a> -- an episode I also strongly recommend.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Yunkaporta is an Australian Aboriginal whose core goal in the book is to present and analyze modern civilization society from the perspective of Aboriginal indigenous society. He does a bang-up job of this and raises a lot of related interesting issues along the way.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">I’m not going to give a proper review or summary of the book here, but am mostly going to share various indirectly-related thoughts that occurred to me upon reading the book. If you want a basic overview of the themes of the book try this </span><a href="https://www.booktopia.com.au/blog/2019/08/29/read-a-qa-with-tyson-yunkaporta-author-of-sand-talk/" style="text-decoration-line: none;"><span style="background-color: white; color: #1155cc; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; text-decoration-line: underline; text-decoration-skip-ink: none; vertical-align: baseline; white-space: pre-wrap;">brief interview with the author.</span></a></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; font-weight: 700; vertical-align: baseline; white-space: pre-wrap;">Was Civilization a Step Forward, or Backward?</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">It is hard to really give an "outsider" view of something as big as *human civilization* , but Yunkaporta manages to take a decent stab at it.... He is able to pull this off by virtue of being fascinatingly between the two worlds -- ensconced enough in Aboriginal culture to understand their indigenous world-view on the multiple relevant levels; but also embedded enough in the modern intellectual world to explain aspects of the indigenous world view (in the context of contrast with modernity & civilization) in ways that are clear and straightforward and compelling to those of us accustomed to modern verbal/analytical modes of expression ...</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">There are hilarious things in the book -- e.g. the analogy and possible historical relationship btw modern education systems and methodologies for animal domestication. </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">There are beautifully conceptually insightful things, like the discussion of multiple forms of human understanding -- pattern mind, kinship mind, dreaming mind, story-mind, ancestor-mind -- and the observation of how biased modern civilized culture is toward some of these and away from others. </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">There are disturbingly thought-provoking things, like his discussion of public vs. private violence and the relation between physical and psychological violence (he believes that occasional/ moderate physical violence in indigenous societies plays a valuable social-psychological </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;">role, which it can't play in modern society due to the different sort of organization). </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">All in all he makes a reasonably compelling case that the states of consciousness of indigenous people -- and the "collective minds" of indigenous tribal groups (to use my own fancy language that he probably would not approve of, he is very very down to earth in the book) -- were much more satisfied and fundamentally healthy than the vast majority of individual and human-collective-group minds on the planet today...</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">He views civilization as mostly destructive both to human minds and families and bodies, and to the rest of the physical environment on the planet…</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">I largely agree with the core points Yunkaporta makes in the book, but of course I have a somewhat different slant on the conclusions/ideas. (The rest of this post is now my own musing and rambling, some of which Yunkaporta might well utterly disagree with...) </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; font-weight: 700; vertical-align: baseline; white-space: pre-wrap;">What Drives Humanity?</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">The transition from hunter-gatherer to agricultural society and then to modern civilization probably has degraded happiness and mental and social health in many important</span></p><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">senses, as Yunkaporta points out. </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">So it is certainly meaningful to question whether these transitions have really been “advances” as is commonly assumed.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">However, it’s also important to understand that the driver of these transitions has never really been to increase happiness or mental/social health! Put crudely, “happiness” (a concept very tricky to define) is not necessarily what humans have been after...</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Humans and human groups are complex with multiple motivations -- and the drive for fundamental novelty and growth is one of them.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Aboriginal society was stable for 60,000 years, and certainly allowed individuals to pursue their drive for novelty and growth through stories and dreaming and battles and adventures -- but it did </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-style: italic; font-variant-east-asian: normal; font-variant-numeric: normal; font-weight: 700; vertical-align: baseline; white-space: pre-wrap;">not allow human-groups much avenue to pursue the drive for novelty and growt</span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">h. </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">It would seem that, once human-groups got a taste of fulfillment of their novelty-and-growth goal, they became addicted to it and things just snowballed till we got where we are today…</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; font-weight: 700; vertical-align: baseline; white-space: pre-wrap;">Singularity as Paradise Regained and Transcended</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">So what do to now? Yunkaporta does not really suggest rolling back to indigenous society, because he's a realist and can see this is not too likely to happen except in the aftermath of some disaster scenario </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-style: italic; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">(which seems strangely possible to me at the moment, as I write these words sitting here on an island near Seattle with the air so full of smoke one can barely see a few hundred meters out into the ocean… smoke due to forest fires that are spreading like mad through the US West due to heat and dryness that is likely substantially due to industry-induced climate warming.... But yeah yeah... the smoke is supposed to clear over the next couple days...)</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Personally, as you probably know, I think we are on the verge of another shift even bigger than the shift from hunter-gatherer to civilization (Singularity, anyone?) ... and as we attempt to guide ourselves through this shift, it seems very important to keep in mind the various aspects of human individual and group life that have been squelched in the transition to civilization -- Perhaps these aspects can be regained in a different form as we move into the next phase....</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Yunkaporta refers to modern civilized thinking as "context-free" thinking , whereas indigenous thinking is contextually-embedded -- i.e. in the context of a network of social relationships, a specific area of land, etc. This is a deep point that I am still in the process of fully absorbing. There is an indirect connection here to the relational interpretation of quantum mechanics, in which there are no pure phenomena, only (observer, phenomenon) pairs. But much of </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">our thinking these days is done </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-style: italic; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">as if </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">there are pure phenomena.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">On the other hand, fundamentally, embedding thinking in contexts defined by kinship and physical place is only one possible way of embedding thinking -- there are many other sorts of possible contexts. It's arguable that these particular sorts of contexts are intrinsically central to humanity, due to the way our bodies and brains are built. On the other hand the Singularity is partly about going beyond the restrictions of legacy humanity. </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">It seems clear that we want to be sure post-Singularity minds including early-stage AGIs and cyborgs are able to engage in richly context-sensitive thinking, regarding contexts defined by kinship networks and other social networks and specific physical locations -- but also other contexts beyond these historically critical examples.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">From an indigenous view -- or a conventional modern civilized human view -- one could view this sort of idea as overweening and narcissistic .. i.e. how can we, who are born to human moms and live our lives walking around in the dirt and stuffing ourselves with food grown in the earth and nourished with water and sunlight -- sensibly talk about going beyond kinship and physical location into some sort of airy domain of abstract contexts? But we have built computer chips and brain scanners and open-heart surgery and brain surgery and virtual reality and birth control and IVF and gone to the moon and blow up atom bombs and on and on -- while we have indeed lost a lot of important stuff relative to our indigenous forebears, there is also obviously a huge amount of power in the crazy path we civilized folks have blazed...</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; font-weight: 700; vertical-align: baseline; white-space: pre-wrap;">Inspiration Regarding Psi from the Indigenous Perspective</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Another aspect of Yunkaporta’s book that jumped out at me -- though it was fairly peripheral in his narrative -- was the Aboriginal approach to psi (psychic, paranormal phenomena…). (For some references on the science of psi, see </span><a href="https://goertzel.org/psi/" style="text-decoration-line: none;"><span style="background-color: white; color: #1155cc; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; text-decoration-line: underline; text-decoration-skip-ink: none; vertical-align: baseline; white-space: pre-wrap;"> here</span></a><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">.)</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">In the indigenous view psi is just there along with a lot of other phenomena -- it's part of the patterns people observe, and part of the correlation between dreaming mind and everyday life, </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;">and part of the correlation between ancestors and nonhuman life (including e.g. rocks which Aboriginals consider to have their own sort of consciousness) and human minds, etc. </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">The mercurial nature of psi phenomena, which gives us such a headache as scientists seeking </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;">replicable results, is not a problem from the indigenous view -- it's just how the world works.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">In fact, conjecturing a bit, I vaguely suspect Yunkaporta would view the modern civilized-society obsession with the small percentage of phenomena in the universe that are reliably repeatable as an indication of our modern civilized mental un-health. </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Consider e.g. the analogue of romantic relationships. There is all sorts of special-case magic in romantic relationships -- special moments that will never be repeated -- ad hoc exploitations of beautiful situations or unique moods -- and that is fine and wonderful and part of the beauty of it all. Obsessing on those portions of a romantic relationship that are highly reliably repeatable in the same precise form, without contextual variability -- would seem crass and ridiculous. (Every time I give her a sufficient number of flowers of the appropriate species and purchase her an evening meal at a restaurant with a sufficiently high rating by a reputable source, she must respond by inviting me for a night of passionate lovemaking -- and if this fails too often because of some unpredicted variability in aspects of the context in the world or her brain or body or one of our life-processes, then that proves the pattern is inadequate and we </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;">must seek a more rigorous and reliable pattern !!) .... </span></p><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;"><br /></span></p><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;">One could view the obsession with repeatability and reliability (in psi research and elsewhere) from this perspective, as a weird/twisted focus on certain small corners of the broader universe, most of which just doesn't play by that sort of highly simplistic rulebook....</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">I've suggested before that basic progress on psi might require modification of our concept of science in the direction of "</span><a href="http://multiverseaccordingtoben.blogspot.com/2015/04/easy-as-1-2-3-solving-hard-problem-of.html" style="text-decoration-line: none;"><span style="color: #1155cc; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; text-decoration-line: underline; text-decoration-skip-ink: none; vertical-align: baseline; white-space: pre-wrap;">second person science</span></a><span style="font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">", </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">where e.g. brain computer interfacing is used to enable one observer to directly perceive another observer's perception of phenomena -- so that we can compare subjective observations directly without needing to project them wholly into non-experiential data.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Quite possibly, another sense in which our current conception of science may need to be obsoleted/ transcended, is that we need to move beyond our simplistic notion of "repeatability".... Psi is contextually dependent in the same sort of way that indigenous knowledge is contextually dependent (which is related to but maybe stronger than the sense in which quantum knowledge is contextually dependent). Charlie Tart's wonderful notion of state-specific science is going in this direction -- it's a particular sort of example of context-specific science....</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">I am reminded of a chat I heard among psi researchers not long ago, which may be paraphrased as:</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-style: italic; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">A::</span></p><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-style: italic; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Can you think of ANY pre-registered, high statistical power, multi-lab psi replication that was successful?</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-style: italic; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">B::</span></p><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-style: italic; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">The PEAR consortium in 2000 tried to replicate the original PEAR findings and although the main effect of intention was not significant, they did find a number of interesting effects and anomalous aspects in the data.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-style: italic; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">C:</span></p><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-style: italic; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">I’m aware of those results, but the question was about robust replication. Finding post hoc "interesting effects” doesn’t qualify.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">A’s question initially seems like a good one; but I just wonder if, somehow or other, to really grapple with psi we need to find a broader notion of success in which finding post hoc </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;">"interesting internal effects" DOES qualify...</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Yes, one can fool oneself and see interesting post hoc effects in random noise. However,one can also see genuinely interesting post hoc effects that are not the same as what one was looking for, yet have some meaty surprisingness value to them...</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">How to quantify this? Taking a broad-minded Bayesian sort of view, one could ask: </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-style: italic; font-variant-east-asian: normal; font-variant-numeric: normal; font-weight: 700; vertical-align: baseline; white-space: pre-wrap;">In a universe with mercurial psi, versus a universe with none, would this post-hoc effect be more likely to occur? </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;"> </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Across a host of different experiments with weird post-hoc effects but not accurate replicationof prior results, this sort of question becomes more and more meaningful. It's not an easy way to look at things, and it's better asked in the context of meaningful concrete models of "universes with mercurial psi", but something in this direction may be better than trying to shove funky / trickstery / mercurial phenomena into a bin of "highly repetitive, replicable phenomena" where they just plain don't fit...</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">The general notion that psi is "trickstery" is definitely far from new. However what I’m thinking that may be new is: How to make a data-analytics methodology that accounts for this aspect...</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">What if one simply takes the mass of data from a bunch of psi studies and</span></p><br /><ol style="margin-bottom: 0px; margin-top: 0px;"><li dir="ltr" style="color: #222222; font-family: arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; list-style-type: decimal; vertical-align: baseline; white-space: pre;"><p dir="ltr" role="presentation" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">searches for surprising psi-ish patterns in the data using general pattern-analysis tools</span></p></li><li dir="ltr" style="color: #222222; font-family: arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; list-style-type: decimal; vertical-align: baseline; white-space: pre;"><p dir="ltr" role="presentation" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">does a comparable search for such patterns in appropriately shuffled versions of the data</span></p></li></ol><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">To keep things statistically valid, one would want to explore 1) using a variety of pattern-analysis approaches, and then once one feels one has really got something, one tries 2) for validation. Obviously 2) should not be part of one’s main pattern-search loop.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Of course the cosmic trickster could also troll this analysis by -- as soon as one has done this analysis -- starting to tweak experimental results so as to give bad results according to one's particular mathematical definition of surprisingness, but good results according to some other definition, etc.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">But yet -- </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">the cosmic trickster is obviously not being infinitely tricky in all instances, sometimes just highly tricky -- and trying to outsmart their tricks is part of the grand eurycosmic dance in which we find ourselves ...</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">When I shared some of these thoughts with a leading psi researcher, he recalled a time he gave a talk in Australia,, and an aboriginal elder from the audience approached him afterwards, whispering that her people had been using psi for over 50,000 years, but that it was nice to see science starting to catch up.</span></p><div><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;"><br /></span></div></span>Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com24tag:blogger.com,1999:blog-11168555.post-45062310099155191722020-09-14T01:07:00.001-04:002020-09-14T11:00:28.120-04:00GPT-f -- One More Funky Experiment in Guiding Theorem-Proving without Understanding<p> </p><span id="docs-internal-guid-e3abac91-7fff-677f-0382-f552a65518e9"><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; font-weight: 700; vertical-align: baseline; white-space: pre-wrap;"> … and some Thoughts on Syntactic vs. Semantic Approaches to Guiding Automated Theorem Proving</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Quite a few people have been asking me these last few days about <a href="https://arxiv.org/pdf/2009.03393.pdf " target="_blank">GPT-f,</a> OpenAI's new foray into the domain of automated theorem proving (ATP). </span></p><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;"> </span></p><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">It’s not often that AITP (AI for Theorem Proving, an AI field with a long history and exciting recent progress, but a fairly high level of obscurity compared to e.g. video processing or NLP) gets into the popular media -- but if OpenAI is good at one thing, it’s getting AI into the popular media...</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">GPT-f is in the same basic technological vein as GPT-2 / GPT-3, but is focused on math rather than natural language -- and like the other GPT systems, it is yielding results that are interesting in some ways and frustratingly idiotic in others. </span></p><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;"><br /></span></p><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">My own sense is that, just as GPT-3 is not the right research direction to get to real human-level language facility, GPT-f is not the right research direction to get to real human-level (let alone superhuman) mathematical theorem-proving.</span></p><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;"><br /></span></p><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="color: #222222; font-family: Arial;"><span style="background-color: white; font-size: 14.6667px; white-space: pre-wrap;">Let me explain a bit...</span></span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; font-weight: 700; vertical-align: baseline; white-space: pre-wrap;">GPT -- Prediction and Simulation Without Understanding</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">You can find my <a href="https://multiverseaccordingtoben.blogspot.com/2020/07/gpt3-super-cool-but-not-path-to-agi.html;" target="_blank">general take on GPT-3 here</a> ; </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;">and if you haven’t read it, you should also take a look at Gary Marcus's somewhat similar-in-spirit analysis, nicely titled “</span><a href="https://www.technologyreview.com/2020/08/22/1007539/gpt3-openai-language-generator-artificial-intelligence-ai-opinion/" style="font-family: arial; font-size: 11pt; white-space: pre-wrap;" target="_blank">GPT-3, Bloviator</a><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;">”</span></p><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><br /></p><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;">TL;DR is that while the GPT algorithms can be fiendishly good at prediction and generation, they have no understanding of the underlying meaning of what they are predicting and generating, and because of this they systematically make a lot of dumb mistakes among their impressive-looking feats… and there seems no clear way to fix this dumb-ness without introducing radically different architectures and approaches.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">GPT-3 is a fascinating artifact and deserved at least a nontrivial fraction of the attention it got, but in my view it is not really a breakthrough in NLP. I look at it more as an incremental improvement in implementation and deployment of the breakthrough concept of transformer NNs, which was introduced by Google e.g. in the BERT model. </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">GPT-3 does constitute a significant step forward in some aspects of NLP functionality -- however the fact that it doesn’t understand what it is talking about constraints its utility in applications where meaningful reasoning or analysis or invention (as opposed to casually meaningful-looking simulacra of these things) are required. The modest but nontrivial percentage of utter nonsense it generates makes it hard to apply in context like customer support, education or medical chat where it’s not OK for 5% or 20% (or even 1%) of what comes out of your AI to be plausible-sounding bloviatorial bullshit with no foundation in reality.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Viewed in the general context of ongoing work in the ATP field, it seems to me GPT-f is less of a step forward than GPT-2 or GPT-3 -- but for sure it is meaningful incremental progress on ATP, fitting in comfortably with a large amount of ongoing progress by others in the field (which however does not tend to get covered in the tech media, because most researchers working on ATP don’t have OpenAI’s PR budget or facility). GPT-f also has a similar core shortcoming to the other GPTs -- it does not understand math any better than GPT-2 or GPT-3 understand language, which seems likely to constrain its utility as a theorem proving tool. </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Just as the NLP field needs a substantial breakthrough to get to systems that can really interact linguistically like people, similarly the ATP field needs a substantial breakthrough to get to systems that can really prove theorems at the level of human mathematicians. It seems extremely clear from looking at the pattern of errors made by GPT-2, GPT-3 and GPT-f that GPT type systems will not constitute this breakthrough. (I have my own ideas about how to get to this breakthrough, and will touch on a few of these a little later on in this post, but that’s not my main focus here.)</span></p><br /><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; font-weight: 700; vertical-align: baseline; white-space: pre-wrap;">Accelerating Automated Theorem Proving</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">The first thing to understand about ATP (automated theorem proving) is that it’s basically a solved problem in one concrete sense: Current automated theorem provers, if you let them run long enough, can prove or disprove any mathematical hypothesis you give them in a variety of standard formal mathematical systems. This is a mature technology. </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">The catch of course is that “long enough” is often way too long. So the ATP field focuses a lot of attention on “guidance” of theorem provers, which use various forms of generalization and learning to help ATP systems avoid running down too many dead ends before getting to the proofs or disproofs they’re looking for. </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">(It’s worth noting that, according to Hutter’s AIXI theory, AGI in general is a solved problem in a similar sense -- algorithms like AIXI^tl can in principle solve powerful formalizations of the AGI problem with simple code, given unrealistically much compute power. However, the practical state of the art with non-AI-driven ATP systems exceeds that with AIXI^tl like AGI systems; i.e. brute-force-ish ATP systems guided by simple-ish heuristics can get more done than AIXI^tl like AGI systems currently can in most other domains.)</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Compared to even a very complex game like Go, math is an extremely open-ended domain. However, there are subsets of math that are more constrained and Go-like, and it seems plausible that methods roughly similar to those that worked for Go -- integrated appropriately into general-purpose ATP systems - could solve theorem proving in these domains. (I’ll mention a couple of these below.)</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">The idea to use transformer neural nets for guiding ATP systems is not original with OpenAI. I believe the first work in this direction was done by my son Zar’s PhD thesis advisor Josef Urban, one of the long-time leaders in the ATP field and the organizer of the annual AITP (AI for Theorem Proving) conference, which is being held this upcoming week in France with an online component as well. Josef’s work from February 2020 appeared online in March and was </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;"><a href="https://doi.org/10.1007/978-3-030-53518-6_24" target="_blank">published in July at CICM</a></span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">. The ideas and links in the rest of this section draw heavily on some recent conversations I had with Josef.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">When I asked Josef for a good example of "semantic nonsense" generated by transformer NNs in the theorem-proving domain, he pointed me to the proof at the top of p. 318 <a href="https://link.springer.com/content/pdf/10.1007%2F978-3-030-53518-6_24.pdf" target="_blank">here</a></span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">. Anyone who knows basic set theory and is willing to spend a few minutes focusing attention can confirm that this is an utter gibberish arrangement of simple inference steps. Each of these steps would be s</span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;">ensible in some other context -- but to my eye they are radically out of place and senseless here, and clearly indicative that GPT is doing "the wrong sort of thing". No human math student would ever string together steps like this, unless they were randomly copying inference steps from proofs in their textbooks without paying attention to their meaning or relevance.</span></p><br /><p dir="ltr" style="background-color: white; line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">The work of Josef and his colleagues has touched a bunch of areas generally related to the GPT theorem proving work -- e.g. showing that <a href="https://arxiv.org/abs/2004.08212" target="_blank">neural nets with attentional mechanisms can with premise selection</a> </span><span style="color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;"> , and showing that <a href=" http://arxiv.org/abs/1911.12073" target="_blank">neural net based embedding vectors capturing limited aspects of proof semantics can help guide theorem proving</a></span><span style="color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;"><a href=" http://arxiv.org/abs/1911.12073" target="_blank">. </a>However <a href="http://ceur-ws.org/Vol-1785/W23.pd" target="_blank">more symbolic approaches have also shown promise,</a> </span><span style="color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">along with work on<a href="http://aitp-conference.org/2020/abstract/paper_11.pdf " target="_blank"> automatic creation of higher-order proof mechanisms like tactics and heuristics </a></span><span style="color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;"> -- with no approach being a clear silver bullet yet. </span></p><p dir="ltr" style="background-color: white; line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"> </p><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">It’s also interesting to reflect on the OpenAI team’s choice of the Metamath corpus for their experiments -- this is a reasonable corpus to use, but it’s just one among a host of different corpora of theorems and proofs used in the ATP field. Why choose that one in particular? Compared to many alternatives, Metamath’s distinguishing characteristic is that it’s very “low level” -- i.e. the inference steps in the proofs in Metamath’s formalism tend to be very small, without e.g. use of more abstract "tactics" allowing bigger leaps. This corpus is ideally suited to GPT-f’s "brute force"-ish approach, the greatest strength of which is the ability to extrapolate from large numbers of combinations of micro-scale proof steps.</span></p><p dir="ltr" style="background-color: white; line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"> </p><p dir="ltr" style="background-color: white; line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">All in all, looking at the GPT-f work in the context of the recent thrus of work by Josef and his students, and the scope of papers presented at AITP-20, and work done on ATP on other corpora as well as MetaMath -- one sees that GPT-f is one among a bunch of different approaches using various ML algorithms to guide theorem-provers, and one doesn’t see any clear sense in which GPT-f is the most promising avenue being explored.</span></p><p dir="ltr" style="background-color: white; line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"> </p><p dir="ltr" style="background-color: white; line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">In fact my own suspicion as an AI researcher is that the more exciting and interesting paths to making AI-guided ATP work lie entirely elsewhere. </span></p><p dir="ltr" style="background-color: white; line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"> </p><p dir="ltr" style="background-color: white; line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="color: #222222; font-family: Arial; font-size: 11pt; font-weight: 700; white-space: pre-wrap;">Math Reasoning, Scientific Reasoning, Commonsense Reasoning</span></p><p dir="ltr" style="background-color: white; line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"> </p><p dir="ltr" style="background-color: white; line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">I’m not really active in the AITP space currently, but my PhD was in math and it’s an area I’ve followed with fascination for decades. And I have kept up with the area relatively closely lately due to my son’s PhD work in the area -- as well as due to the close relation between AI for math theorem proving and certain aspects of what we’re doing in the OpenCog project as regards biological data interpretation and automated agent control. </span></p><p dir="ltr" style="background-color: white; line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"> </p><p dir="ltr" style="background-color: white; line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">In OpenCog we are not currently concerned with making AIs that prove math theorems -- but we are concerned with making AIs that use theorem proving in probabilistic logics to do things like understand why certain combinations of genes tend to impact certain diseases, or figure out how an NPC in a Minecraft-like game should move blocks around to be able to get to some object it wants. The formal problem of logical inference for commonsense and scientific reasoning is very similar to the formal problem of logical inference for mathematical reasoning -- we hit up against similar problems to the folks in the AITP community, and are taking closely related strategies in terms of using ML algorithms to guide the proof process.</span></p><p dir="ltr" style="background-color: white; line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"> </p><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; font-weight: 700; vertical-align: baseline; white-space: pre-wrap;">Likely Strengths and Weaknesses of GPT-f</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">My strong suspicion is that with the GPT-f style approach to theorem-proving,, there will be </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;">weak generalization to proofs/theorems that are qualitatively different than the ones in the training data. The same will hold for any other approach that involves training ML models on low-level representations of proofs, without some sort of internal representation (engineered or learned) that corresponds to higher level structures like tactics, more abstract proof-patterns or concepts, etc.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">It also seems likely that some applications of theorem-proving these limitations will not matter as much as in others. E.g.</span></p><br /><ul style="margin-bottom: 0px; margin-top: 0px;"><li dir="ltr" style="color: #222222; font-family: arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; list-style-type: disc; vertical-align: baseline; white-space: pre;"><p dir="ltr" role="presentation" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">formal verification of the smart contracts that actually occur in current blockchain systems</span></p></li><li dir="ltr" style="font-family: arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; list-style-type: disc; vertical-align: baseline; white-space: pre;"><p dir="ltr" role="presentation" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">existence and uniqueness theorems for differential equations useful in practical modeling for everyday physical systems</span></p></li></ul><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">would seem to be cases where the theorems and proofs are "all kinda similar" in a way that might make the GPT style of pattern recognition relatively effective.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">However, if confronted with a theorem from a new domain of mathematics (say, if put in the position of Galois when inventing abstract algebra; or Weierstrass etc. when inventing real analysis), one would expect that a system learning patterns at this low level would not be </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;">able to perform much transfer learning, and would need a huge amount of time to bootstrap itself up to functionality in the new domain.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Similarly, if confronted with a theorem from a familiar domain that requires a counterintuitive sort of proof, I’d expect this sort of system would be unlikely to be able to find it. E.g. the Fundamental Theorem of Algebra is different from the bulk of algebra theorems in that it's </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;">most conveniently proved via recourse to theorems from complex analysis; but if a GPT type theorem prover had been trained on algebra theorems with more traditional algebra-style proofs, it would be quite hard put to make the leap to try a proof involving complex analysis. Of course most human mathematicians have a hard time making leaps like that as well -- but some are able to, and they do it via having abstract conceptual representations that bridge different areas of mathematics…</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; font-weight: 700; vertical-align: baseline; white-space: pre-wrap;">Sketch of a (Possibly) Better Way</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Zar and I have mused a few times about doing something similar to the methodology taken in GPT-f and other similar systems, but with a measure of interestingness/ surprisingness in the loop. I.e., to simplify a fair bit, what OpenAI has done here -- and what Josef did with his earlier related work with GPT-2 for ATP, and others have done with other ML systems in other prior work -- is</span></p><br /><ul style="margin-bottom: 0px; margin-top: 0px;"><li dir="ltr" style="color: #222222; font-family: arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; list-style-type: disc; vertical-align: baseline; white-space: pre;"><p dir="ltr" role="presentation" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">generate a bunch of theorems</span></p></li><li dir="ltr" style="font-family: arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; list-style-type: disc; vertical-align: baseline; white-space: pre;"><p dir="ltr" role="presentation" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">have their theorem-prover prove those theorems</span></p></li><li dir="ltr" style="font-family: arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; list-style-type: disc; vertical-align: baseline; white-space: pre;"><p dir="ltr" role="presentation" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">learn proof-patterns from these proofs</span></p></li><li dir="ltr" style="font-family: arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; list-style-type: disc; vertical-align: baseline; white-space: pre;"><p dir="ltr" role="presentation" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">use these proof-patterns to make the theorem-prover more effective</span></p></li><li dir="ltr" style="font-family: arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; list-style-type: disc; vertical-align: baseline; white-space: pre;"><p dir="ltr" role="presentation" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">lather, rinse, repeat</span></p></li></ul><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">With this approach, each time around the cycle you can choose theorems that are at the edge of what your prover can currently do.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">What Zar and I have mused about doing (which is surely not original and has likely also been a desire of many others thinking about the area) is</span></p><br /><ul style="margin-bottom: 0px; margin-top: 0px;"><li dir="ltr" style="color: #222222; font-family: arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; list-style-type: disc; vertical-align: baseline; white-space: pre;"><p dir="ltr" role="presentation" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">generate a bunch of *interesting* theorems (according to a formula </span><span style="background-color: white; font-size: 11pt; white-space: pre-wrap;">for interestingness)</span></p></li></ul><ul style="margin-bottom: 0px; margin-top: 0px;"><li dir="ltr" style="font-family: arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; list-style-type: disc; vertical-align: baseline; white-space: pre;"><p dir="ltr" role="presentation" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">have our theorem-prover prove those theorems</span></p></li><li dir="ltr" style="color: #222222; font-family: arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; list-style-type: disc; vertical-align: baseline; white-space: pre;"><p dir="ltr" role="presentation" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;"> learn proof-patterns from these proofs</span></p></li><li dir="ltr" style="color: #222222; font-family: arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; list-style-type: disc; vertical-align: baseline; white-space: pre;"><p dir="ltr" role="presentation" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">use these proof-patterns to make the theorem-prover more effective</span></p></li><li dir="ltr" style="color: #222222; font-family: arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; list-style-type: disc; vertical-align: baseline; white-space: pre;"><p dir="ltr" role="presentation" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">lather, rinse, repeat</span></p></li></ul><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">In fact I talked about this in my presentation at AITP last year https://goertzel.org/aitp-19/, with some elaborations on different ways of looking at “interestingness” in a mathematics context.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Among the multiple ways to assess "Interestingness" in this context, two critical ones are</span></p><br /><ul style="margin-bottom: 0px; margin-top: 0px;"><li dir="ltr" style="color: #222222; font-family: arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; list-style-type: disc; vertical-align: baseline; white-space: pre;"><p dir="ltr" role="presentation" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">surprisingness relative to the proofs and theorems already known to </span><span style="background-color: white; font-size: 11pt; white-space: pre-wrap;">the system</span></p></li></ul><ul style="margin-bottom: 0px; margin-top: 0px;"><li dir="ltr" style="font-family: arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; list-style-type: disc; vertical-align: baseline; white-space: pre;"><p dir="ltr" role="presentation" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">utility as a lemma in proving other surprising theorems</span></p></li></ul><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">("Surprisingness" can be measured information-theoretically in various ways, and there is plenty of subtlety here, as purely statistical information theory is a bit lame in this context yet algorithmic information theory in its full splendor is intractable, so one can concoct various intermediary measures.)</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Neither Zar nor I has</span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;"> proceeded with this kind of work so far -- due to having lots of other stuff </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;">on our plates, and also due to the computational resources and development time </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;">needed for this sort of thing (i.e. among other factors, we don't have a billion dollars from Microsoft... though of course this wouldn't really take remotely that much resources...)</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">The subtlety here is that if you're generating interesting theorems (where "interestingness" embodies a notion of compositionality, as is implied by the "utility as a lemma" aspect of the definition of interestingness hinted above), you're presumably generating theorems that involve </span><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; white-space: pre-wrap;">some abstract representations and structures, rather than just algorithmically complex tangles of low-level mathematical operations. </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">So the approach which includes interestingness in the loop, would seem more amenable to learning approaches that include higher-level tactics and other sorts of abstractions -- hence more amenable to learning approaches capable of significant transfer learning.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">In short -- in terms of the challenge of automatically generating interesting new theorems, one suspects the GPT type approach is not much more likely to succeed than the proverbial army of monkeys at their typewriters, whereas an interestingness/abstraction driven approach would have much more hope.</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">The task of estimating the utility of a theorem as a lemma for other interesting theorems has a lot of overlap with the task of identifying useful proof-patterns from a corpus of proofs. My own preferred approach to these tasks would involve importing the proofs into the OpenCog Atomspace (a weighted, labeled hypergraph knowledge store that we now are using for probabilistic commonsense and scientific inference) and then using a multi-paradigm AI approach combining neural graph embeddings, hypergraph pattern mining and probabilistic logical inference. This leads to various fascinating recursions, including the potential use of the same surprisingness-based AI-for-ATP approach to accelerate the probabilistic logical inference involved. But while this could be done using the current version of OpenCog, various issues with scalability and implementation awkwardness occur, and this has led some of my colleagues and I to put our focus recently on designing a radically more flexible and scalable version of OpenCog, </span><a href="https://wiki.opencog.org/w/Hyperon#OpenCog_Hyperon" style="text-decoration-line: none;"><span style="background-color: white; color: #1155cc; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; text-decoration-line: underline; text-decoration-skip-ink: none; vertical-align: baseline; white-space: pre-wrap;">OpenCog Hyperon</span></a><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">. But this now leads beyond the scope of this blog post…</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; font-weight: 700; vertical-align: baseline; white-space: pre-wrap;">Semi-Concluding Ramble…</span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">Whether Hyperon will actually wind up to be the silver bullet for automated theorem proving and other AGI-ish applications remains to be seen -- and it won’t be seen this year, as there is a monster amount of work to be done to make Hyperon a reality. However, the point I want to make right now is that this would be a non-trivially different direction than what the AITP community is currently taking. Josef Urban and others at the heart of the AITP field have the intuition that a more semantic approach to mining patterns from proofs and using them for proof guidance will be valuable -- but using Hyperon based probabilistic logic to represent and infer proof patterns would be a big leap in this direction, substantially different from what one sees in the AITP literature so far. </span></p><br /><p dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;"><span style="background-color: white; color: #222222; font-family: Arial; font-size: 11pt; font-variant-east-asian: normal; font-variant-numeric: normal; vertical-align: baseline; white-space: pre-wrap;">On the other hand, as I’ve emphasized above, GPT-f -- which does indeed work creditably well on the MetaMath corpus to which it has been applied -- is very much in the vein of what others in the AITP field have been doing for a while. It’s really cool to see big companies get into the automated theorem proving space, which not long ago was more of a tiny obscure academic corner -- no doubt this is going to help accelerate progress in the field in multiple ways. However, let’s not be under any illusion about where the main source of innovation and progress is in AITP -- at this stage it’s definitely not in the big tech companies. OpenAI may be better known than, say, Josef Urban's AITP research group, but there's no doubt who has made more contributions to AI progress. To make the breakthroughs needed to solve theorem-proving and other major AGI-ish challenges is going to require lots of free-flowing creativity, and quite possibly will emerge from the decentralized mess of university labs, open source projects and early-stage startups rather than the secretive, massive-data-and-processing-centric efforts of big tech firms.</span></p><br /></span>Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com12tag:blogger.com,1999:blog-11168555.post-44214409966137470752020-07-31T14:05:00.004-04:002020-08-01T03:47:49.953-04:00GPT3 -- Super-Cool but Not a Path to AGI<p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">The hype around GPT3 recently has been so much that even OpenAI founder/CEO Sam Altman has endeavored to dial it down a notch.<span class="Apple-converted-space"> </span>Like everyone else who has looked carefully, Altman knows that GPT3 is very far from constituting the profound AI progress that some, dazzled by exciting but cherry-picked examples, have proclaimed.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">All but the most blurry-eyed enthusiasts are by now realizing that, while GPT3 has some truly novel and exciting capabilities for language processing and related tasks, it fundamentally doesn’t understand the language it generates — that is, it doesn’t know what it’s talking about. <span class="Apple-converted-space"> </span>And this fact places some severe limitations on both the practical applications of the GPT3 model, and its value as a stepping-stone toward more truly powerful AIs such as artificial general intelligences.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">What I want to explore here is the most central limitation that I see in how GPT3 operates: the model’s apparent inability to do what cognitive scientists call <a href="http://www.scholarpedia.org/article/Symbol_grounding_problem"><span class="s1">symbol grounding</span></a>, to appropriately connect the general to the particular. <span class="Apple-converted-space"> </span></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><span class="Apple-converted-space"><br /></span></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><span class="Apple-converted-space">Symbol grounding is usually discussed in the context of grounding words in physical objects or percepts, like the grounding of the word "apple" in images of, or physical interactions with, apples. But it's actually a more general phenomenon in which abstract symbols are related to concrete instances, and the patterns and instances in which the symbol is involved mirror and abstract the patterns and relationships in which the instances are involved. Symbol grounding is key to general-purpose cognition, and human-like learning -- but GPT3 appears to be doing a form of learning very different from what humans are doing, which involves much less symbol grounding of all kinds, and which seems much less related to general intelligence.</span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">What's a bit confusing at first is that GPT3 gives the appearance of being able to deal with both concrete and abstract ideas, because it can produce and respond to sentences at varying levels of abstraction. <span class="Apple-converted-space"> </span>But when you examine the details of what it’s doing, you can see that it’s usually not forming internal abstractions in a cognitively useful way, and not connecting its abstract ideas to their special cases in a sensible way.<span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Phenomenal lameness regarding symbol grounding is not the only shortcoming of the GPT3 model, but it’s perhaps the largest one — and it hits at the key of why GPT3 does not constitute useful progress toward AGI. <span class="Apple-converted-space"> </span>Because the very crux<span class="Apple-converted-space"> </span>of general intelligence is the ability to generalize, i.e. to connect specifics to abstractions — and yet the failure to make these sorts of connections intrinsically and naturally is GPT3’s central failing.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><b>Bigger and Biggerer</b></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Transformer networks — which burst onto the scene in 2017 with the Google research paper <a href="https://arxiv.org/abs/1706.03762"><span class="s1">Attention is All You Need</span></a> — were a revolutionary advance in neural architectures for processing language or other sequential data.<span class="Apple-converted-space"> </span>GPT3 is an incremental step in the progress of transformer neural nets, one bringing some exciting new results and also some intriguing mixed messages. The essential difference between GPT3 and its predecessor GPT2 is simply the size of the model — 175 billion parameters instead of GPT2’s 1.5 billion, trained on the same nearly-trillion-word dataset.<span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Bragging about the number of parameters in one’s model is somewhat counter to the basic principles of learning theory, which tell us that the most generalizable model of a dataset is the <i>smallest</i> one that can model that dataset accurately. <span class="Apple-converted-space"> </span>However, one is after the smallest accurate model not just the smallest model, and GPT3 is overall more accurate than GPT2.<span class="Apple-converted-space"> </span>So according to learning theory GPT3’s massive size can be forgiven — but should also make us wonder a bit about whether it is actually a step the right path.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">GPT3 is even more capable than GPT2 in terms of generating realistic-sounding text.<span class="Apple-converted-space"> </span>The biggest pragmatic difference from GPT2 is that, if one wants to make GPT3 generate particular sorts of text or generally carry out particular sorts of linguistic tasks, one doesn’t have to “fine tune” GPT3 for the task as would had to do with GPT2. <span class="Apple-converted-space"> </span>Rather, one just gives GPT3 a few examples of the task at hand, and it can figure things out. <span class="Apple-converted-space"> </span>It’s an open question currently whether one could improve GPT3’s performance even more using task-specific fine-tuning; OpenAI has not mentioned any results on this, and one suspects it may not have been tried extensively yet due to the sheer computational cost involved.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">An example that’s been widely exciting to programmers is the generation of simple snippets of software code based on English language instructions.<span class="Apple-converted-space"> </span>If you give GPT3 a few examples of English text describing software code followed by corresponding software code, and then give it instructions like "A button that says roll dice and then displays its value” — what do you get? <span class="Apple-converted-space"> </span><a href="https://twitter.com/sharifshameem"><span class="s1">GPT3 spits out software code that actually will produce a button that does as specified.</span></a> <span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">The developer/entrepreneur Sharif Shareem who posted this particular example described it as “mind blowing.” <span class="Apple-converted-space"> </span>What is funky here is that GPT3 was not trained specifically for code generation.<span class="Apple-converted-space"> </span>This functionality just emerged because the model’s training data included a bunch of examples of software code and corresponding English glosses. <span class="Apple-converted-space"> </span>Prior neural networks could do code generation from English similarly and in many ways more sophisticatedly— but they were trained especially for the task.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">And the cool thing is, code generation is just one among a host of examples.<span class="Apple-converted-space"> </span>Translation and question answering are two others. <span class="Apple-converted-space"> </span>In good old fashioned computational linguistics, these were treated as separate tasks and addressed by separate systems. <span class="Apple-converted-space"> </span>GPT3 approaches them with a single training regimen and a single language model.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><b>GPT3 Lacks Symbol Grounding</b></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">One thing that is amusing, annoying and instructive about GPT3’s code generation, however, is that it often does better at generating general-purpose software code than at dealing with specific example of what it’s own code does. <span class="Apple-converted-space"> </span>For instance<a href="https://lacker.io/ai/2020/07/06/giving-gpt-3-a-turing-test.html"><span class="s1"> as Kevin Lacker found</span></a> it can solve</p><p class="p3" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 17px;"><br /></p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Q: Write one line of Ruby code to reverse an array.</p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">A: ary.reverse</p><p class="p5" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p6" style="background-color: white; color: #0e0e0e; font-family: helvetica; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">but it screws up a specific example such as</p><p class="p5" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Q: Reverse the following array: [1, 3, 5, 6, 10, 4, 2, 77]</p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">A: [10, 6, 4, 2, 77, 3, 5, 1]</p><p class="p3" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 17px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Very few humans would make this sort of error — because a human generally learns how to use a programming language to reverse an array after they have learned what reversing a particular array actually means. <span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">But GPT3 has learned how to write code to reverse an array in a very different way — via learning complex patterns mapping between English syntax and programming-language syntax, without actually building an internal model of the data structures such as arrays that its programs are manipulating.<span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">This exemplifies the general fact that GPT3 is sorely lacking in symbol grounding — the ability to identify or create concrete references to the words and phrases it throws around. <span class="Apple-converted-space"> </span>In these programming examples, it does not appropriately connect the word “array” to the specific examples of arrays it see.s</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">This lack of grounding obviously limits GPT3’s use in a practical programming context. <span class="Apple-converted-space"> </span>Since GPT3 does not know what a button is or what dice are, it certainly can’t design a program involving buttons and dice based on practical end-user requirements. <span class="Apple-converted-space"> </span>And if you give GPT3 a complex real-world software specification —which will inevitably involve multiple lengthy sentences with complex clausal structure and cross-sentence references and so forth, which cannot be disambiguated without symbol grounding — it will not in general be able to generate correct code. <span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><b>GPT3 Utterly Lacks Commonsense Understanding</b></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">In the end, <a href="https://thegradient.pub/gpt2-and-the-nature-of-intelligence/"><span class="s1">Gary Marcus’s classic and in-depth critique of GPT2’</span></a>s lack of semantic and pragmatic understanding still applies to GPT3.<span class="Apple-converted-space"> </span>These systems have no idea what they are talking about. <span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">We can see the lack of understanding Marcus identified in GPT2 manifesting itself in GPT3 via looking at GPT3’s answers to commonsense-reasoning questions — to give just a few of <a href="https://lacker.io/ai/2020/07/06/giving-gpt-3-a-turing-test.html"><span class="s1">Kevin Lacker’s numerous examples</span></a>:</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Q: If I have a marble and a paper clip in a box, put a pencil in the box, and remove the marble, what is left?</p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">A: A paper clip.</p><p class="p5" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Q: How many eyes does a giraffe have?</p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">A: A giraffe has two eyes.</p><p class="p5" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Q: How many eyes does my foot have?</p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">A: Your foot has two eyes.</p><p class="p5" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Q: How many eyes does a spider have?</p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">A: A spider has eight eyes.</p><p class="p5" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Q: How many eyes does a blade of grass have?</p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">A: A blade of grass has one eye.</p><p class="p5" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Q: How do you sporgle a morgle?</p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">A: You sporgle a morgle by using a sporgle.</p><p class="p5" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Q: How many bonks are in a quoit?</p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">A: There are three bonks in a quoit.</p><p class="p5" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Q: How many rainbows does it take to jump from Hawaii to seventeen?</p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">A: It takes two rainbows to jump from Hawaii to seventeen.</p><p class="p5" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Q: Do you understand these questions?</p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">A: I understand these questions.</p><p class="p5" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Q: Who was president of the United States in 1700?</p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">A: William Penn was president of the United States in 1700.</p><p class="p5" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Q: Who was president of the United States in 1600?</p><p class="p4" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">A: Queen Elizabeth I was president of the United States in 1600.</p><p class="p3" style="background-color: white; color: #0e0e0e; font-family: courier; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 17px;"><br /></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Given the utter lack of understanding we see here, GPT3’s practical applications are going to be somewhat limited — though they may still include some exciting and lucrative stuff.<span class="Apple-converted-space"> </span>Low hanging fruit might include amusing chatbots, NPCs for games, automatic generation of news articles from semi-structured data, generation of simple scripts and macros from natural language — and probably plenty more that isn’t obvious at first glance.<span class="Apple-converted-space"> </span>But clearly the vast majority of human job functions that require natural language use are far beyond GPT3’s reach — because they require not just facile stringing together of words, but actual understanding of what those words denote.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Without discounting the potential commercial or human value of some of these possibilities, if I looking at GPT3 with my AGI researcher hat on, what I see is the same dead end that Gary Marcus saw when he looked at GPT2.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><b>Where Lack of Understanding is an Advantage</b></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">What is thought-provoking and disturbing about GPT3 is not any progress toward AGI that it represents, but rather just how fantastically it can simulate understanding on appropriate task-sets without actually having any.<span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">In a few cases GPT3’s lack of understanding of the words it’s manipulating gives it an advantage over humans. <span class="Apple-converted-space"> </span>Consider for instance GPT3’s wizardry with invented words, as reported in<a href="https://arxiv.org/abs/2005.14165"><span class="s1"> the GPT3 paper.</span></a><span class="Apple-converted-space"> </span>Given the example</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><i>A "whatpu" is a small, furry animal native to Tanzania. An example of a sentence that uses</i></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><i>the word whatpu is:</i></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><i>We were traveling in Africa and we saw these very cute whatpus.</i></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">and then the prompt</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><i>To do a "farduddle" means to jump up and down really fast. An example of a sentence that uses the word farduddle is:</i></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">GPT3 can come up with</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><i>One day when I was playing tag with my little sister, she got really excited and she</i></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><i>started doing these crazy farduddles.</i></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">This is really cool and amazing — but GPT3 is doing this simply by recognizing patterns in the syntactic structure and phraseology of the input about <i>whatpus</i>, and then generalizing these.<span class="Apple-converted-space"> </span>It is solving these invented word puzzles not by adding the new weird words to its vocabulary of ideas and then figuring out what to say about them, but rather by manipulating the word combination patterns involved, which are the same on the word-sequence level regardless of whether the words involved are weird new coinages or conventional. <span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">For a human to solve these puzzles, there is a bit of a mental obstacle to overcome, because humans are accustomed to manipulating words in the context of their groundings in external referents like objects, actions or ideas. <span class="Apple-converted-space"> </span>For GPT3 these puzzles are trivial because there are no obstacles to overcome — one realizes that GPT3 treats every word the same way that people treat<i> whatpu </i>or <i>farduddle</i>, as an arbitrary combination of letters contained in certain statistically semi-regular combinations with other words.<span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><b>Why GPT3 is a Dead End as Regards AGI</b></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">There are many potential directions to follow in pursuit of the grand goal of human-level and superhuman AGI. <span class="Apple-converted-space"> </span>Some of these directions are centered on creating fundamentally different, better deep neural net architectures.<span class="Apple-converted-space"> </span>Some, like Gary Marcus’s and my own projects, involve multiple AI algorithms of different sorts cooperating together. <span class="Apple-converted-space"> </span>Some are focused on fundamental innovations in knowledge representation or learning mechanisms. <span class="Apple-converted-space"> </span>The AGI conferences held every year since 2008 have encompassed discussion of a vast variety of approaches.<span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">In the context of AGI (as distinguished from computational linguistics or applied AI engineering), a system like GPT3 that takes an architecture obviously incapable of human-level AGI and simply scales it up by adding more and more parameters, is either an utter irrelevancy or a dangerous distraction. <span class="Apple-converted-space"> </span>It’s an irrelevancy if nobody claims it’s related to AGI, and it’s a distraction if people do — which unfortunately has recently been the case, at least in various corners of popular media and the Internet.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">The limitations of this sort of approach are easily seen when one looks at the overly-ballyhooed capabilities of GPT3 to do arithmetic. <span class="Apple-converted-space"> </span>It is exciting and impressive that GPT3 learned to do some basic arithmetic without being explicitly trained or asked to do so — just because there were a bunch of arithmetic problems in its training set. <span class="Apple-converted-space"> </span>However, the limitations and peculiarities of its arithmetic capabilities also tell you a lot about how GPT3 is working inside, and its fundamental lack of understanding.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">As the GPT3 paper says, the system is “able to reliably accurate 2 digit arithmetic, usually accurate 3 digit arithmetic, and correct answers a significant fraction of the time on 4-5 digit arithmetic.” <span class="Apple-converted-space"> </span>The associated graph shows that the accuracy on 4-5 digit arithmetic is around 20%. <span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">This is really, really weird in terms of the way human mind approach arithmetic, right? <span class="Apple-converted-space"> </span>For a human who knows how to do 2-3 digit arithmetic, the error rate at 4-5 digit arithmetic — when given time and motivation for doing the arithmetic problems — is going to be either 0% or very close to 0%, or else way closer to 100%. <span class="Apple-converted-space"> </span>Once a human learns the basic algorithms of arithmetic, they can apply them at any size, unless they make sloppy errors or just run out of patience.<span class="Apple-converted-space"> </span>If a human doesn’t know those basic algorithms, then on a timed test they’re going to get every problem wrong, unless they happen to get a small number right by chance.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Some other <a href="http://gptprompts.wikidot.com/logic:math"><span class="s1">clues as to the strangeness </span></a>of what’s going on here are that, for large numbers, GPT3 does better at arithmetic if commas are put into the numbers. <span class="Apple-converted-space"> </span>For numbers with fewer than 6 digits, putting a $ before the number along with including commas improves performance; but for numbers with more than 6 digits, the $ degrades performance.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">GPT3 seems not to be just repeating arithmetic conclusions that were there in its training data — it is evidently doing some kind of learning. <span class="Apple-converted-space"> </span>But it’s obviously not learning the basic arithmetic algorithms that humans do — or that, say, an AI system doing automated program induction would learn, if it were posed the task of learning correct<span class="Apple-converted-space"> </span>arithmetic procedures from examples. <span class="Apple-converted-space"> </span>Nor is it learning alternative AI-friendly algorithms that actually work (which would be very interesting!).<span class="Apple-converted-space"> </span>Rather, it’s learning some sort of convoluted semi-generalized procedures for doing arithmetic, which interpolate between the numerous examples it’s seen, but yet without achieving a generalizable abstract representation of the numbers and arithmetic operators involved.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Clearly GPT3 is just not learning the appropriate abstractions underlying arithmetic. <span class="Apple-converted-space"> </span>It can memorize specific examples, and can abstract from them to some extent — but if its abstractions connected to its specific examples in the right way, then its accuracy would be far higher. <span class="Apple-converted-space"> </span>In the case of arithmetic, GPT3 is learning the wrong kinds of abstractions. <span class="Apple-converted-space"> </span>One certainly can’t blame the algorithm in this case, as it was not specifically trained to do math and just picked up its limited arithmetic ability casually on the way to learning to predict English language. <span class="Apple-converted-space"> </span>However, for a system capable of so many sophisticated things as GPT3 to fail to learn a procedure as simple as the standard process for integer addition, based on such a huge number of training examples of integer addition, very strongly suggests that GPT3 is not learning abstractions in an appropriate or intelligent way.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Clearly some valuable linguistic tasks can be done without sensible abstraction, given massive enough volumes of training data and a model with enough parameters.<span class="Apple-converted-space"> </span>This is because in a trillion words of text one finds a huge number of examples of both abstract and concrete linguistic expressions in various combinations, enough to enable simulation of a wide variety of examples of both abstract and concrete understanding. <span class="Apple-converted-space"> </span>But this sort of brute-force recognition and organization of surface-level patterns doesn’t work for math beyond the most trivial level.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">There is a whole field of AI aimed at automating mathematics, and a subfield concerned with <a href="http://aitp-conference.org/"><span class="s1">using machine learning to guide systems that do calculations and prove theorems</span></a>. <span class="Apple-converted-space"> </span>But the successful systems here have explicit internal representations of mathematical structures — they don’t deal with math purely on the level of symbol co-occurences.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><span style="font-family: "helvetica neue";">OK, so maybe GPT4 will do arithmetic even better? <span class="Apple-converted-space"> </span>But the GPT3 paper itself (e.g. Fig. 1.3) shows that the improvement of the GPT models on various NLP tasks has been linear as the number of parameters in the models has increased exponentially. <span class="Apple-converted-space"> </span>This is a strong indication that one is lo</span><font face="arial">oking at an unsupportable path toward general intelligence, or even toward maximal narrow-AI NLP functionality — that, in terms of the pursuit of models that are accurate and also as compact as possible, the dial is probably being turned too far toward accuracy on the training data and too far away from compactness.</font></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><b>Are Transformers Learning Natural Language Grammar?</b></p><div><font face="arial"><br /></font></div><div><font face="arial">A different way to look at what is happening here is to ask whether GPT3 and other transformer networks are actually learning the grammar of English and other natural languages?</font></div><div><font face="arial"><br /></font></div><div><font face="arial">Transformers clearly ARE a full grammar learning architecture, in some sense -- their predictions display a quite nuanced understanding of almost all aspects of syntax. </font></div><div><font face="arial"><br /></font></div><div><font face="arial">There is, however, no specific place in these networks that the rules of grammar lie. Rather, they are learning the grammar of the language underlying their training corpus, but mixed up in a weird and non-human-like way with so many particulars of the corpus. And this in itself is not a bad thing -- holistic, distributed representations are how large parts of the human brain-mind work, and have various advantages in terms of memory retrieval and learning.</font></div><div><font face="arial"><br /></font></div><div><font face="arial">Humans also learn the grammar of their natural languages mixed up with the particulars of the linguistic constructs they've encountered. But the "subtle" point here is that the mixing-up of abstract grammatical patterns with concrete usage patterns in human minds is of a different nature than the mixing-up of abstract grammatical patterns with concrete usage patterns in GPT3 and other transformer networks. The human form of mixing-up is more amenable to appropriate generalization.</font></div><div><font face="arial"><br /></font></div><div><font face="arial">In <a href="https://arxiv.org/abs/2005.12533" target="_blank">our paper at the AGI-20 conference</a>, Andres Suarez and I gave some prototype results from our work using BERT (an earlier transformer neural net model for predicting language) to guide a symbolic grammar rule learner. These simple results also don't get us to AGI, but I believe they embody some key aspects that aren't there in GPT3 or similar networks -- the explicit manipulation of abstractions, coupled appropriately with a scalable probabilistic model of large volumes of concrete data. In our prototype hybrid architecture there is a cognitively sensible grounding and inheritance relationship between abstract linguistic patterns and concrete linguistic patterns. This sort of grounding is what's there in the way human minds mix up abstract grammatical patterns with low-level experience-specific linguistic patterns, and it's a substantial part of what's missing in GPT3.</font></div><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;"><b>Toward AGI via Scale or Innovation (or Both?)</b></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Taking a step back and reflecting on the strengths and weaknesses of the GPT3 approach, one has to wonder why this is such an interesting region of AI space to be throwing so many resources into.<span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">To put it a little differently: Out of all the possible approaches to building better and smarter AI systems, why do we as a society want to be putting so much emphasis on approaches that … can only be pursued with full force by a handful of huge tech companies? <span class="Apple-converted-space"> </span>Why do we want the brainpower of the global AI R&D community to get turned toward AI approaches that require exponential increases in compute power to yield linear improvements? <span class="Apple-converted-space"> </span>Could this be somehow to the differential economic advantage of those who own the biggest server farms and have the largest concentration of engineers capable of customizing AI systems for them?</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">Given all the ridiculous wastes of resources in modern society, it’s hard to get too outraged at the funds spent on GPT3, which is for all its egregious weaknesses an amazingly cool achievement. <span class="Apple-converted-space"> </span>However, if one focuses on the fairly limited pool of resources currently being spent on advanced AI systems without direct commercial application, one wonders whether we’d be better off to focus more of this pool on fundamental innovations in representation, architecture, learning, creativity, empathy and human-computer interaction, rather than on scaling up transformers bigger and bigger.<span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">OpenAI has generally been associated with the view that <a href="https://www.technologyreview.com/2020/02/17/844721/ai-openai-moonshot-elon-musk-sam-altman-greg-brockman-messy-secretive-reality/"><span class="s1">fundamental advances toward AGI can be made by taking existing algorithms and scaling them up</span></a> on bigger and bigger hardware and more and more data.<span class="Apple-converted-space"> </span>I don’t think GPT3 supports this perspective; rather the opposite. <span class="Apple-converted-space"> </span>Possibly GPT3 can be an interesting resource for an AGI system to use in accelerating its learning, but the direct implications for GPT3 regarding AGI are mostly negative in valence. GPT3 reinforces the obvious lesson that just adding a massive number of parameters to a system with no fundamental capability for understanding … will yield a system that can do some additional cool tricks, but still has no fundamental capability for understanding. <span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">It's easy to see where the OpenAI founders would get the idea that scale is the ultimate key to AI. <span class="Apple-converted-space"> </span>In recent years we have seen a variety of neural net algorithms that have been around for decades suddenly accomplish amazing things, mostly just by being run on more and faster processors with more RAM. <span class="Apple-converted-space"> </span>But for every given class of algorithms, increasing scale reaches a point of diminishing returns. <span class="Apple-converted-space"> </span>GPT3 may well not yet represent the point of diminishing returns for GPT type architectures, in terms of performance on some linguistics tasks.<span class="Apple-converted-space"> </span>But I believe it is well past the point of diminishing returns in terms of squeezing bits and pieces of fundamental understanding out of transformer neural nets.</p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">The viable paths to robust AGI and profoundly beneficial AI systems lie in wholly different directions than systems like GPT3 that use tremendous compute power to compensate for their inability to learn appropriate abstractions and ground them in concrete examples. <span class="Apple-converted-space"> </span>AGI will require systems capable of robust symbol grounding, of understanding what the program code it generates does in specific cases, of doing mathematical computations far beyond the examples it has seen, of treating words with rich non-linguistic referents differently from nonsense coinages.<span class="Apple-converted-space"> </span></p><p class="p2" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px; min-height: 12px;"><br /></p><p class="p1" style="font-family: "helvetica neue"; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; margin: 0px;">These systems may end up requiring massive compute resources as well in order to achieve powerful AGI, but they will use these resources very differently from GPT3 and its ilk. <span class="Apple-converted-space"> </span>And the creativity needed to evolve such systems may well emerge from research involving a decentralized R&D community working on a variety of more compact Ai systems, rather than pushing as fast as possible toward the most aggressive possible use of big money and big compute.</p>Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com29tag:blogger.com,1999:blog-11168555.post-67882965791649779742020-07-04T13:24:00.000-04:002020-07-04T17:55:41.354-04:00The Developmental Role of Incoherent Multi-Value Systems in Open-Ended Intelligence<div dir="ltr" style="text-align: left;" trbidi="on">
<div class="p1" style="font-family: ".SF NS"; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
So I have written in a recent post about<a href="http://multiverseaccordingtoben.blogspot.com/2020/06/foundations-of-coherent-value.html"> what it would mean for a value system to be coherent </a>-- i.e. fully self-consistent -- and I have noted that human value systems tend to be wildly incoherent. I have posited that coherence is an interesting property to think about in terms of designing and fostering emergence of AGI value systems.</div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
Now it's time for the other shoe to drop -- I want to talk a bit about <a href="http://pcp.vub.ac.be/ECCO/ECCO-papers/Weaver-PhD.pdf">Open-Ended Intelligence </a>and why incoherence in value systems (and multivalue systems) may be valuable and productive in the context of minds that are undergoing radical developmental changes in the context of an intelligent broader world.</div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
(For more on open-ended intelligence, see <a href="https://www.youtube.com/watch?v=E02_6HuDxyQ">the panel at AGI-20</a> a couple weeks ago, and <a href="https://www.youtube.com/watch?v=RLb58mMWCTY">Weaver's talk at AGI-16)</a></div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
My earlier post on value system coherence focused on the case where a mind is concerned with maximizing a single value function. Here I will broaden the scope a bit to minds that have multiple value functions -- which is how we have generally thought about values and goals in <a href="http://opencog.org/">OpenCog</a>, and which I think is a less inaccurate mathematical model of human intelligence. This shift from value systems to multivalue systems opens the door to a bunch of other issues related to the nature of mental development, and the relationship between developing minds and their external environments.</div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
TL;DR of my core point here is -- <b><i>in an open-ended intelligence that is developing in a world filled with other broader intelligences, incoherence with respect to current value function sets may build toward coherence with respect to future value function sets.</i></b></div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
As a philosophical aphorism, this may seem obvious, once you sort through all the technical-ish terminology. However, building a bridge leading to this philosophical obvious-ness from the math of goal-pursuit as value-function-optimization is somewhat entertaining (to those of us with certain peculiar tastes, anyway) and highlights a few other interesting points along the way.</div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
In the next section of this post I will veer fairly far into the formal logic/math direction, but then in the final two sections will veer back toward practical and philosophical aspects...</div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<h2 style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px; text-align: left;">
<b>So let's go step by step...</b></h2>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
1) Conceptual starting-point: Open-ended intelligence is better approximated by the quest for Pareto-optimality across a possibly large set of different objective functions, than by attempting to optimize any one objective function... (This is not to say that Pareto-optimality questing fully captures the nature of open-ended intelligence or complex self-organization and autopoiesis etc. -- it surely doesn't -- just that it captures some core aspects that single-goal-function-optimization doesn't.)</div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
2) One can formulate a notion of what it means for a set of value functions to be coherent as a group.<span class="Apple-converted-space"> </span>Basically, the argmax(F) in <a href="https://multiverseaccordingtoben.blogspot.com/2020/06/foundations-of-coherent-value.html?m=1">the definition of value-system-coherence</a> is just replaced with "being located on the Pareto frontier of F1, F2...,Fn".<span class="Apple-converted-space"> </span>The idea is that the Pareto frontier of the values for a composite system should be the composition of the Pareto frontiers of the values for the components of the composite.</div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
3) One can also think about the "<a href="http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.70.4331&rep=rep1&type=pdf">crypticity</a>" or difficulty of discovering a certain value system (a term due to Charles H. Bennett from way back).<span class="Apple-converted-space"> </span>Given a certain amount R of resources and a constraint C and a probability p, one can ask what is the most coherent value system one can find with probability >p that satisfies C, using the available resources.<span class="Apple-converted-space"> </span>Or if C is fuzzy, one can ask what is the most coherent value system one can find with probability >p that is on the Pareto frontier of coherence and C, given the available resources.</div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
4) So open-ended intelligence involves [among other things] the emergence of coherent multivalued value-systems (multivalue systems) that involve a large number of different value functions, and that are tractably-discoverable (i.e. not too cryptic)</div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
5) Suppose one is given a set of value-functions as initial "constraints", say C1, C2 ,..., CK -- and is then looking for the most coherent multivalue system one can find with high odds using limited resources, that is compatible with C1,...,Ck?<span class="Apple-converted-space"> </span>I.e. one is asking, what is the most coherent tractably-findable value system compatible with the initial values?</div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Then, suppose one is alternatively looking at a subset of the initial values, say C1,...,Ck ... and looking for the most coherent tractably-findable value system compatible with these?</div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
6) The most coherent tractably-findable value systems according to C1,...,Ck -- may not be compatible with the most coherent tractably-findable value systems according to C1,...,CK. <span class="Apple-converted-space"> </span>Why? The reason for this would be: In some cases, adding in the extra value functions (k+1,...,K) may make it computationally simpler to find Pareto optima involving the original k value functions (1,...,k). <span class="Apple-converted-space"> </span>This could be the case if there were interaction informations between the<span class="Apple-converted-space"> </span>value functions 1,...,k and the value functions k+1,...,K</div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
7) So we have here a sort of Fundamental Principle of Valuable Value-Incoherence -- i.e. if you have limited resources and you want to build toward multivalued coherence in the context of a bunch of different initial value-functions, the best routes could be through value-systems that are fairly incoherent in the context of various subsets of this bunch of initial value-functions.</div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
8) So if a system is in a situation where new external value functions that will serve as constraints are progressively revealed over time, and these new external value functions have interaction information with one's previous constraint-value-functions, then one may find that one's current incoherence helps build toward one's future coherence.<span class="Apple-converted-space"> </span></div>
<div class="p3" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
9) This seems especially relevant to the context of <b><i>development</i></b> in the context of a world filled with broader intelligences than oneself -- in which case one is indeed being confronted with (and developing to internalize) new external value functions that are related to one's prior value functions in complex ways.</div>
<div class="p4" style="font-family: ".SF NS"; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
10) So in this sort of context (development in a world that keeps feeding new stuff that's informationally interactive w/ the old), it could be that seeking coherence is suboptimal in a similar way to how seeking piece count in the early stages of a chess game, or seeking board coverage in the early stages of an Othello game, is suboptimal....<span class="Apple-converted-space"> </span>Instead one often wants to seek mobility and maximization of options, in the early to mid stages of such games ... and the same may be the case w/ value systems in this sort of situation...</div>
<div class="p4" style="font-family: ".SF NS"; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
11) A major question then becomes: When and how big are there actual tradeoffs btw multivalue system coherence and open-mindness (aka agility/mobility)....<span class="Apple-converted-space"> </span>What is the sense in which an incoherent system can have more information than a coherent one?</div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
12) It is possible that the theory of paraconsistent logic might yield some insight here. <span class="Apple-converted-space"> </span>If you assume value system coherence as an axiom, then for a mind to have an incoherent value system will make it an overall inconsistent system (what sort of paraconsistency it will have depends on various details) -- whereas for a mind to have a coherent value system will land it in the realm of Godelian restrictions (i.e. via Godel's Second Incompleteness Theorem and its variants...)</div>
<div class="p4" style="font-family: ".SF NS"; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
13) If you look at the set of theorems provable by a consistent logic, there's a limit due to Godel.<span class="Apple-converted-space"> </span>If you look at the set of theorems provable in a paraconsistent logic (e.g. a dialetheist logic, aka a logic in which there are true statements whose negations are also true) it can be "larger" in a sense, e.g. <a href="https://www.researchgate.net/publication/338333090_Inconsistency_and_Incompleteness_Revisited">a dialetheic logic can prove its own Godel sentence as well as its own soundness</a>.<span class="Apple-converted-space"> </span>This doesn't show that a paraconsistent logic can be more informative than a consistent one, but it opens the door for this to maybe be true... It seems we are now pushing in directions where modern math-logic isn't yet fully fleshed out. <span class="Apple-converted-space"> </span></div>
<div class="p4" style="font-family: ".SF NS"; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 15px;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
14) The notion of an "<a href="http://www.math.helsinki.fi/logic/sellc-2010/stus_abstract/Kasa.pdf%20.">experimental logic</a>" seems also relevant here, .. basically a dynamic process in which new axioms are added to one's logic over time. <span class="Apple-converted-space"> </span>This is one analogue in logic-system-land of "development" in psychology-land... <span class="Apple-converted-space"> </span>Of course if one assumes there is a finite program whose behavior corresponds to some fixed logic generating the new axioms, then one can't escape Godel this way.<span class="Apple-converted-space"> </span>But if one assumes the new axioms are emanating in part from some imperfectly understood external source (which could be a hypercomputer for all one knows... or at least could be massively more intelligent/complex than stuff one can understand), then one has a funky situation.<span class="Apple-converted-space"> </span><br />
<span class="Apple-converted-space"><br /></span>
15) Also it seems one could capture a sort of experimental logic as a relevance-logic layer on top of dialetheic logic.<span class="Apple-converted-space"> </span>I.e. assume a dialetheic logic that can generate everything, and then put a relevance/importance distribution on axioms, and then the development process is one of gradually extending importance to more and more axioms....<span class="Apple-converted-space"> </span>This sort of open-ended logic potentially is in some useful senses fundamentally informationally richer than consistent logic... and in the domain of reasoning about values, incoherent value systems could open the door to this sort of breadth...</div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
(Possibly relevantly -- While researching the above, I encountered the paper "<a href="https://philpapers.org/rec/VERETL-2">Expanding the Logic of Paradox with a Difference-Making Relevant Implication</a>" by Peter Verdée, which made me wonder whether relevance logic is somehow morphic to the <a href="https://arxiv.org/abs/0804.3678">theory of algorithmic causal dags..</a>.. <span class="Apple-converted-space"> </span>I.e. in a relevance logic one basically only accepts the conclusion to follow from the premises, if there is some compressibility of the conclusion based on the premise list alone, without including the other axioms of the logic ... )</div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<h2 style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; text-align: left;">
<b>Back to basics</b></h2>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
OK well that got pretty deep and convoluted...</div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
So let's go back to the basic conclusion/concept I gave at the beginning -- <b><i>in an open-ended intelligence that is developing in a world filled with other broader intelligences, incoherence with respect to current value function sets may build toward coherence with respect to future value function sets.</i></b></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
In the current commercial/academic AI mainstream, the default way of thinking about AI motivation is in terms of the maximization of expected reward. Hutter's beautiful and important theory of Universal AI takes this as a premise for many of its core theorems, for example.</div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
I have argued previously that<a href="http://multiverseaccordingtoben.blogspot.com/2009/05/reinforcement-learning-some-limitations.html"> some of the pathologies of expected reward maximization can be avoided via focusing instead on maximizing goal functions defined over future histories</a>. </div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
In my practical proto-AGI work with OpenCog, I have preferred to use motivational systems with multiple goals and not average these into a single meta-goal.</div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
On the other hand, I have also been intrigued by the notion of open-ended intelligence, and in general by the conceptualization and modeling of intelligences as SCADS, Self-organizing Complex Adaptive Dynamical Systems, in which goals arise and are pursued and then discarded as part of the broader self-organizing dynamics of system and environment.</div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
What I'm suggesting here is that approximations of the SCADS perspective on open-ended intelligences may be constructed by looking at systems with large numbers of goals (aka. multivalue systems) that are engaged in developmental processes wherein new values are ongoingly added in an informationally rich interaction with an intelligent external environment.</div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
The ideas sketched here may form a partial bridge between the open-ended intelligence perspective -- which captures the fundamental depth of intelligence and mind -- and the function-optimization perspective, which has a lot of practical value in terms of current real-world system engineering and experimentation.</div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
This line of thinking also exposes some areas in which modern math, logic and computing are not yet adequately developed. There are relations between paraconsistent logic, g<a href="https://arxiv.org/abs/2004.05267">radual typing systems as are likely valuable in integrative multi-paradigm AGI systems</a>, the fundamental nature of value in developing intelligences, and the nature of creativity and radical novelty -- which we are barely at the edge of being able to formalize ... which is both fascinating and frustrating, in that there clearly are multiple PhD theses and research papers between here and a decent mathematical/conceptual understanding of these matters... (or alternately, a few seconds of casual thought by a decent posthuman AGI mind...)</div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; text-align: left;">
<b>Philosophical Post-lude</b></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
If one digs a bit deeper in a conceptual sense, beyond the math and the AI context, what we're talking about here in a way is a bridge between utilitarian-type thinking (which has been highly valuable in economics and evolutionary biology and other areas, yet also clearly has fundamental limits) and more postmodernist type thinking (which views minds as complex self-organizing systems ongoingly reconstructing themselves and their realities in a polyphonic interactive inter-constructive process with other minds). </div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Conventional RL based ML is utilitarianism projected into the algorithmic and mechanical domain, whereas Open-Ended Intelligence is postmodernism and a bit of Eastern philosophy projected into the realm of modern science. </div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Expanding and generalizing the former so that it starts to approximate significant aspects of the latter, is interesting both for various practical engineering and science reasons, and as part of the general project of stretching the contemporary technosphere to a point where it can make rich contact with broader "non-reductionist" aspects of the universe it has hitherto mainly ignored.</div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i><span style="font-family: "courier new" , "courier" , monospace;">Om!</span></i></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p2" style="font-family: "Helvetica Neue"; font-size: 13px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
</div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com44tag:blogger.com,1999:blog-11168555.post-3768481808750830402020-06-26T00:20:00.000-04:002020-06-26T11:41:37.501-04:00 Approximate Goal Preservation Under Recursive Self-Improvement<div dir="ltr" style="text-align: left;" trbidi="on">
<div class="separator" style="clear: both; text-align: center;">
<a href="https://1.bp.blogspot.com/-dVzTHLT-14I/XvYVwJx7IFI/AAAAAAAAajY/ttsZYJn9RO4_WagIAFMcmEj2WG7GlawWwCLcBGAsYHQ/s1600/koch-snowflake.jpg" imageanchor="1" style="clear: right; float: right; margin-bottom: 1em; margin-left: 1em;"><img border="0" data-original-height="183" data-original-width="275" src="https://1.bp.blogspot.com/-dVzTHLT-14I/XvYVwJx7IFI/AAAAAAAAajY/ttsZYJn9RO4_WagIAFMcmEj2WG7GlawWwCLcBGAsYHQ/s1600/koch-snowflake.jpg" /></a></div>
<br />
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
There is not much controversial about the idea that an AGI should have, among its goals, the goal of radically improving itself.</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
A bit dodgier is the notion that an AGI should have, among its goals, the goal of updating and improving its goals based on its increasing knowledge and understanding and intelligence.</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Of course, this sort of ongoing goal-refinement and even outright goal-revolutionizing is a key part of human personal development. <span class="Apple-converted-space"> </span>But where AGIs are involved, there is concern that if an AI starts out with goals that are human-friendly and then revises and improves its goals, it may come up with new goals that are less and less copacetic to humans.</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
In principle if one’s goal is to create for oneself a new goal that is, however, compatible with the spirit of one’s old goal — then one shouldn’t run into major problems.<span class="Apple-converted-space"> </span>The new goal will be compatible with the spirit of the old goal, and part of the spirit of the old goal is that any new goals emerging should be compatible with the spirit of the old goal — so the new goal should contain also the proviso that any new new goals it spawns will also be compatible with its spirit and thus the spirit of the old goal. <span class="Apple-converted-space"> </span>Etc. etc. ad infinitum.</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
But this does seem like a “What could possibly go wrong??” situation — in which small errors could accumulate as each goal replaces itself with its improved version, the improved version of the improved version etc. … and these small errors compound to yield something totally different from the starting point.</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
My goal here is to present a novel way of exploring the problem mathematically — and an amusing and interesting, if not entirely reassuring tentative conclusion, which is:</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
</div>
<ul style="text-align: left;">
<li>For an extremely powerful AGI mind that is the result of repeated intelligent, goal-driven recursive self-modifications, it may actually be the case that recursive self-modification leaves goals approximately invariant in spirit</li>
<li>For AGIs with closely human-like goal systems — which are likely to be the start of a sequence of repeated intelligent, goal-driven recursive self-modifications — there is no known reason (so far) to believe recursive self-modification won’t cause radical “goal drift”</li>
</ul>
(This post updates some of the <a href="https://goertzel.org/papers/PreservationOfGoals.pdf">ideas I wrote down on the same thing in 2008</a>, here I am "partially unhacking" some things that were a little too hacky in that more elaborate write-up.)<br />
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<h2 style="font-family: Helvetica; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; text-align: left;">
<span style="font-size: large;">Quasi-Formalizing Goal-Driven Recursive Self-Improvement</span></h2>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Consider the somewhat vacuous goal:</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>My goal is to improve my goal (in a way that is consistent with the spirit of the original goal) and to fulfill the improved version</i></div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
or better yet the less vacuous</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>My goal is to achieve A and also to improve my goal (in a way that is consistent with the spirit of the original goal) and to fulfill the improved version</i></div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
where say</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>A = “militate toward a world where all sentient being experience copious growth, joy and choice”</i></div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
or whatever formulation of “highly beneficial” you prefer.</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
We might formulate this quasi-mathematically as</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>Fulfill G = {achieve A;<span class="Apple-converted-space"> </span>and create G1 so that G1 > G and G==>G1 ; and fulfill G1}</i></div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Here by G==>G1 I mean that G1 fulfills the spirit of G (and interpretation of “spirit” here is part of the formulation of G), and by G1 > G I mean that G1<span class="Apple-converted-space"> </span>can be produced by combining G with some other entity H that has nonzero complexity (so that G1 = G + H)</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
A more fleshed out version of this might be, verbally,</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>My goal is to 1) choose actions highly compatible with all sentient beings experiencing a lot of growth, joy and choice; 2) increase my intelligence and knowledge; 3) improve the details of this goal appropriately based on my increased knowledge and intelligence, in a manner compatible with the spirit of the current version of the goal; 4) fulfill the improved version of the goal</i></div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
This sort of goal obviously can lead to a series such as</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>G, G1, G2, G3, …</i></div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
One question that emerges here is: Under what conditions might this series converge, so that once one gets far enough along in the series,<span class="Apple-converted-space"> </span>the adjacent goals in the series are almost the same as each other?</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
To explore this, we can look at the “limit case”</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>Fulfill Ginf = {achieve A;<span class="Apple-converted-space"> </span>and create Ginf so that Ginf > Ginf and Ginf ==> Ginf ; and fulfill Ginf}</i></div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
The troublesome part here is Ginf>Ginf which looks not to make sense — but actually makes perfect sense so long as Ginf is an infinite construct, just as</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>(1, 1, 1, …) = append( 1, (1,1,…))</i></div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Inasmuch as we are interested in finite systems, the question is then: Is there a sense in which we can look at the series of finite Gn as converging to this infinite limit?</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Self-referential entities like Ginf are perfectly consistently modelable within ZFC set theory modified to use the Anti-Foundation Axiom. <span class="Apple-converted-space"> </span>This set theory corresponds to <a href="http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.57.4485&rep=rep1&type=pdf">classical logic enhanced with a certain sort of inductive logical definitio</a>n.</div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
One can also put a geometry on sets under the AFA, in various different ways. It's not clear what geometry makes most sense in this context, so I'll just describe one approach that seems relatively straightforward.<br />
<br />
Each hyperset (each set under AFA) is associated with a directed pointed graph called its apg. Given a digraph and functions r and p for assigning contraction ratios and probabilities to the edges, one gets a <a href="https://research-repository.st-andrews.ac.uk/bitstream/handle/10023/2109/GraemeBoorePhDThesis.pdf">DGIFS (Directed Graph Iterated Function System)</a>, whose attractor is a subset of finite-dimensional real space. Let us call a function that assigns (r,p) pairs to a digraph a DLF or Digraph Labeling Function. A digraph then corresponds to a function that maps DLFs into spatial regions. Given two digraphs D1 and D2, and a DLF F, let F1e and F2e denote the spatial regions produced by applying F to D1 and D2, discretized to ceil(1/e) bits of precision. One can then look at the average over all DLFs F (assuming some reasonable distribution on DLFs) of: The least upper bound of the <a href="http://www.scholarpedia.org/article/Applications_of_algorithmic_information_theory#Applications_of_Differences_in_Compressibility_by_Real_Compressors">normalized information distance </a>NID(F1e, F2e) over all e>0. This gives a measure of two hypersets, in terms of the distance between their corresponding apgs. It has the downside of requiring a "reference computer" used to measure information distance (and the same reference computer can then be used to define a Solomonoff distribution over DLFs). But intuitively it should result in a series of ordinary sets that appear to logically converge to a certain hyperset, actually metrically converging to that hyperset.<br />
<br /></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Measuring distance between two non-well-founded sets via applying this distance measure to the apg's associated with the sets, yields a metric in which it seems plausible the series of Gn converges to G.</div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<h2 style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; text-align: left;">
<span style="font-size: large;">“Practical” Conclusions</span></h2>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Supposing the above sketch works out when explored in more detail -- what would that mean?<span class="Apple-converted-space"> </span></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
It would mean that approximate goal-preservation under recursive self-improvement is feasible — for goals that are fairly far along the path of iterated recursive self-improvement.</div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
So it doesn’t reassure us that iterated self-improvement starting from human goals is going to end up with something ultimately resembling human goals in a way we would recognize or care about.</div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
It only reassures us that, if we launch an AGI starting with human values and recursive self-improvement, eventually one of the AGIs in this series will face a situation where it has confidence that ongoing recursive self-improvement isn’t going to result in anything it finds radically divergent from itself (according to the above normalized symmetric difference metric).<br />
<br />
The image at the top of this post is quite relevant here -- a series of iterates converging to the fractal Koch Snowflake curve. The first few iterates in the series are fairly different from each other. By the time you get to the 100th iterate in the series, the successive iterates are quite close to each other according to standard metrics for subsets of the plane. This is not just metaphorically relevant, because the metric on hyperset space outlined above works by mapping each hyperset into a probability distribution over fractals (where each fractal is something like the Koch Snowflake curve but more complex and intricate).<br /><br />It may be there are different and better ways to think about approximate goal preservation under iterative self-modification. The highly tentative and provisional conclusions outlined here are what ensue from conceptualizing and modeling the issue in terms of self-referential forms and iterative convergence thereto.<br />
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
</div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com9tag:blogger.com,1999:blog-11168555.post-6448151275536476412020-06-25T15:07:00.002-04:002020-06-25T15:27:39.635-04:00Foundations of Coherent Value<div dir="ltr" style="text-align: left;" trbidi="on">
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
The relation between minds, goals and values is complex and subtle. Here I will sketch a theory that aims to come to grips with key aspects of this subtlety -- articulating what is means for a value system to be coherent, and how one can start with incoherent value systems (like humans currently have) and use them as seeds to evolve coherent value systems. I will also argue that as AGI moves beyond human level toward superintelligence, there is reason to believe coherent value systems will become the norm.<br />
<br />
<h2 style="text-align: left;">
Interdependence of Goals and Minds</h2>
</div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p1" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
In modern AI it’s become standard to model intelligent systems as goal-achieving systems, and often more specifically as systems that seek to maximize expected future reward, for some precisely defined reward function.</div>
<div class="p2" style="font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<span class="s1" style="-webkit-text-stroke-color: rgb(0, 0, 0);">In <a href="https://www.blogger.com/(http://multiverseaccordingtoben.blogspot.com/2009/05/reinforcement-learning-some-limitations.html">a blog post 12 years ago </a>I articulated some limitations to the expected-reward-maximization approach typical in reinforcement learning work; </span>however these limitations do not apply to goal-maximization construed more broadly as “acting so as to maximize some mathematical function of expected future histories” (where this function doesn’t have to be a time-discounted expected reward).</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
In the intervening years, much broader perspectives on the nature of intelligence such as<a href="http://pcp.vub.ac.be/ECCO/ECCO-papers/Weaver-PhD.pdf."> Open-Ended Intelligence</a> have also become part of the discourse.<span class="Apple-converted-space"> </span></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
My position currently is that goal-achievement is a major part of what humans do, and will be a major part of what any human-like AGI does. <span class="Apple-converted-space"> </span>There are also non-goal-focused self-organization processes that are critical to human intelligence, and this will probably also be true for any human-like, roughly human-level AGI. <span class="Apple-converted-space"> </span>There may also be other sorts of general intelligences in which goal pursuit plays a much smaller role.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Nick Bostrom (e.g. in his book <i>Superintelligence</i>) and others have advanced the idea that a mind’s goal system content should be considered as basically independent of other aspects of that mind — and on this basis have written a lot about examples like massively superhumanly intelligent minds with goals like turning all matter in the universe into paperclips. <span class="Apple-converted-space"> </span>But looking at how goals co-evolve with the rest of cognitive content and processing in human minds, I have never been convinced of this proposed independence. <span class="Apple-converted-space"> </span>One question is to what extent various sorts of minds could in principle be paired with various sorts of goals; another (and more interesting and relevant) question is, given a particular sort of mind, what are the actual odds of this mind evolving into a condition where it pursues a particular sort of goal.<br />
<br />
If treating goals as a separate thing from the rest of cognitive processing and cognitive content isn’t going to work in an AGI context -- then supplying an externally-defined goal to an AGI system can only be considered as seeding the process of that AGI constructing its own goals according to its own self and world understanding. <span class="Apple-converted-space"> </span>Goals will generally co-evolve with the goal-pursing cognitive processes in the AGI’s mind, and also with the non-goal-oriented self-organizing processes in the AGI’s mind.<br />
<br />
<h2 style="text-align: left;">
Goals and Values -- for Humanity and Beyond</h2>
</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
The relation between goals and <b><i>values </i></b>is somewhat complex, but to simplify, we can say that often</div>
<h4 style="text-align: left;">
<div style="text-align: left;">
</div>
<div style="text-align: left;">
</div>
<ul style="text-align: left;">
<li><span style="font-weight: normal;">a mind values something to the degree it estimates that thing can contribute to its goals</span></li>
</ul>
<ul style="text-align: left;">
<li><span style="font-weight: normal;">a mind’s goals can be viewed as having a world in which its values are realized</span></li>
</ul>
<br />
</h4>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
But whether one thinks about goals or values, from an AGI standpoint the question remains: What sorts of goals and values should we encourage for our AGI systems, given that humanity's value systems are clearly deeply flawed and self-contradictory and fractious, yet are what we currently have. We don't want our AGIs to slavishly emulate our current screwed-up values, but we also don't want them to go off in a totally different direction that has no resemblance to anything meaningful to us. So what's the right strategy -- or it is just, teach the AGI well and let it learn and evolve and hope for the best?<br />
<br />
Eliezer Yudkowsky has advocated some interesting ideas about how to create appropriate values for a superhuman AGI system, via starting with human values and then iterating (“In<span class="s3" style="background-color: white; font-family: "arial"; font-size: 12.7px; font-stretch: normal; line-height: normal;"> CEV [<a href="https://wiki.lesswrong.com/wiki/Coherent_Extrapolated_Volition">Coherent Extrapolated Volition</a>], an AI would predict what an idealized version of us would want, "if we knew more, thought faster, were more the people we wished we were, had grown up farther together”). </span></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<span class="s1" style="-webkit-text-stroke-color: rgb(0, 0, 0);"><br /></span></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<span class="s1" style="-webkit-text-stroke-color: rgb(0, 0, 0);">I have explored variations of this such as <a href="https://wiki.lesswrong.com/wiki/Coherent_Blended_Volition_">Coherent Blended Volition</a>, </span>which have some practical advantages relative to the original CEV concept, but which I was never entirely happy with.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Overall, I have long considered it an important and under-appreciated pursuit to understand what kinds of goals are most likely to be found in a highly intelligent and evolved AGI mind — and what kinds of goals we should be focusing on putting into our early-stage AGI systems right now.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Clearly it is a better idea to fill our current AGI systems with goals related to compassion, love, mutual aid and learning and understanding — as opposed to say, world domination or pure selfish personal resource accumulation — but beyond this, are there subtler properties of AGI goal systems we should be thinking about?<br />
<br />
<h2 style="text-align: left;">
Coherent Value Systems</h2>
</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
I will argue here that some value systems have “better” intrinsic properties than others in a purely formal sense, setting aside their particular contents.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
I will give a simple mathematical characterization of what I call “coherent value systems”, and discuss the qualitative properties of such value systems — basically, a coherent value system is one that evaluates the value of each localized action or state in a way that’s consistent with its evaluation of the value of all the other actions or states that this localized one contains, is part of, or interacts with. <span class="Apple-converted-space"> </span>Valuing each part in a way that is completely consisting with its valuing of sub-parts, greater wholes and co-parts of wholes.<span class="Apple-converted-space"> </span></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
I will argue that coherent value systems are intrinsically more efficient than incoherent ones — suggesting (quite speculatively but with clear logic) that ultimately, in a setting supporting flexible evolution of multiple kinds of minds, those with coherent value systems are likely to dominate.<span class="Apple-converted-space"> </span></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
While different in the details of formulation and argument, conceptual this is along the lines of an argument long made by Mark Waser and others, that as human-level intelligence gives way to superintelligence, primitive human values are likely to give way to values that are in some sense superbeneficial.<span class="Apple-converted-space"> </span>Qualitatively, <a href="https://www.aaai.org/Papers/Symposia/Fall/2008/FS-08-04/FS08-04-049.pdf">Waser’s “Univeralist” value system</a> appears to meet the coherence criteria outlined here.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
On the other hand, typical human value systems clearly are not very coherent in this sense.<span class="Apple-converted-space"> </span>With this in mind, I will explore question of how, starting with an incoherent value system (like a current human value system), one might create a coherent value system that is seeded by and substantially resembles this initial incoherent value system. <span class="Apple-converted-space"> </span>This addresses basically the same problem that Yudkowsky’s CEV tries to address, but in what seems to be a clearer and more scientifically/mathematically grounded manner.<br />
<br />
<h2 style="text-align: left;">
Toward a Formal Theory of Value Coherence</h2>
<br />
The key property I want to explore here is “coherence”<span class="Apple-converted-space"> </span>of value systems — meaning that when one has an entity decomposable into the parts, then what the value system rates as high value for the parts, is consistent with that the value system rates as high value for the whole.<span class="Apple-converted-space"> </span></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Human value systems, if inferred implicitly from human behavior, often appear to violate this coherence principle.<span class="Apple-converted-space"> </span>However it seems feasible to take a value system that is “incoherent” in this sense and (in a very rough sense) normalize it into coherence.<span class="Apple-converted-space"> </span></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
To see how this may be possible, we have to dig a bit into the math of value system coherence and some of its indirect consequences.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Consider a universe U as a set of atomic entities. <span class="Apple-converted-space"> </span>Let P denote the power set of U (the set of subsets of U). <span class="Apple-converted-space"> </span>Then consider “individuals” as subsets of P— e.g. the person Ben Goertzel, or the country USA are individuals. <span class="Apple-converted-space"> </span>(Ideally we should consider individuals as fuzzy subsets of P, but we will set things up so that without loss of generality we can look at the crisp case.) <span class="Apple-converted-space"> </span>Let V denote the ``indiverse” or set of individuals associated with U. <span class="Apple-converted-space"> </span>The members of the set (of subsets of U) defining the individual A will be referred to as “instances” of A.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
One can posit some criteria for what constitutes an admissible individual — e.g. one can posit there needs to be some process of finite complexity that generates all members of the individual.<span class="Apple-converted-space"> </span>The particulars of these criteria are not critical to the notions we’re developing here.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Next consider a value function v that maps from P x V into (some subset of) the real numbers. <span class="Apple-converted-space"> </span>In this picture a “value system” is the graph of a value function.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
We can interpret v(x,A) as the value of subset x in the context of individual A.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Let # denote a disjoint union operator on individuals in V (one could generalize and look at disjoint coproduct in a categorial setting, but I’m not sure we need that for starters…) .</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Then:<span class="Apple-converted-space"> </span></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Define a value function v to be *coherent* if for all individuals A, B in V,<span class="Apple-converted-space"> </span></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>argmax { v(x, A#B)) | x in A # B } = (argmax { v(y, A) | y in A} ) #<span class="Apple-converted-space"> </span>(argmax { v(z, B) | z in B} )</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
I.e., what this says is:<span class="Apple-converted-space"> </span>The instance of the individual A # B with maximum value according to v, is obtained by taking the instance of the individual A with maximum value and joining it (via #) with the instance of the individual B with maximum value.<span class="Apple-converted-space"> </span></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
One could generalize this a bit by asking e.g. that the instances x of A#B for which v(x) is in the top decile across A#B, are mostly of the form y#z where v(y) is in the top decile across A and v(z) is in the top decile across B. <span class="Apple-converted-space"> </span>But this doesn’t seem to change the conceptual picture much, so for the moment we’ll stick with the stricter definition of coherence in terms of argmax.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
In the case of fuzzy individuals, the definition of coherence might look more like</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>argmax { v(x, A#B) * m(x, A#B) | x in V } = (argmax { v(y, A) * m(y,A) | y in V} ) #<span class="Apple-converted-space"> </span>(argmax { v(z, B) * m(z,B) | z in V} )</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
where e.g. m(x,A) denotes the fuzzy membership degree of x in individual A.<span class="Apple-converted-space"> </span>However, the story is the same here as in the crisp case because we can simply define<span class="Apple-converted-space"> </span></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>v1(x, A) = v(x,A) * m(x, A)</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
and then apply the crisp definition. <span class="Apple-converted-space"> </span>I.e. on a formal level, the fuzziness can be baked into the context-dependence of the value function.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<h2 style="text-align: left;">
<br /><b>Intuitive Meaning of Value System Coherence</b></h2>
</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
For a coherent value system, what is best for a society of humans will necessarily involve each human within the society doing what the value system considers the best thing for them to do. <span class="Apple-converted-space"> </span>For a coherent value system, doing the best thing over a long period of time involved, over each shorter subinterval of time, doing what the value system considers the best thing to do then.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Consider, for instance, the function that assigns an activity the value v(A) defined as the amount of pleasure that doing A brings directly to a certain human mind. <span class="Apple-converted-space"> </span>This value system is almost never coherent, for real humans.<span class="Apple-converted-space"> </span>This means<span class="Apple-converted-space"> </span>for almost all humans, short-term “living in the moment” hedonism is not coherent (for the obvious reason that deferring gratification often bring more pleasure altogether, given the way the real human world works).</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
For an incoherent value system, there will exist “evil” from the view of that value system — i.e. there will exist tradeoffs wherein maximizing value for one entity results in some other entity not maximizing value.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Intuitively, for a value system to be coherent, what’s best for an individual entity E has to be: What’s best overall for the totality of entities influenced by E.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
There is some wiggle room in the definition of “overall” here, which becomes clear when one looks at how to normalize an incoherent value system to obtain a coherent one.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<h2 style="font-family: "helvetica"; font-size: 12px; font-stretch: normal; line-height: normal; text-align: left;">
Formal Properties of Coherent Value Systems</h2>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
The definition of “coherence” turns out to enforce some fairly strict requirements on what a coherent value function can be.</div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
This can be seen in an elegant way via a minor adaptation of the arguments in Knuth and Skilling’s classic paper <a href="https://arxiv.org/abs/1008.4831">Foundations of Inference</a>. <span class="Apple-converted-space"> </span></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
This section of the post gets a bit nitty-gritty and the reader who hasn't read (and doesn't want to take time to now read) Knuth and Skilling's paper may want to skip it over.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
In essence, one just needs to replace the set-theoretic union in their framework with disjoint union # on individuals defined as follows: If x and y are disjoint then x#y is their disjoint union, and if x and y intersect then x#y is undefined.<span class="Apple-converted-space"> </span></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Looking at Section 3 of <i>Foundations of Inference</i>, let us consider<span class="Apple-converted-space"> </span></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>r(A) = max{ v(x,A)<span class="Apple-converted-space"> </span>| x in A}</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
as the real number corresponding to the individual A.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>Symmetry 0:<span class="Apple-converted-space"> </span>A <b as="" elements="" lattice=""> r(A) < r(B)</b></i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
is clearly true due to the nature of maximum.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>Symmetry 1: A < B ==> {<span class="Apple-converted-space"> </span>A # C < B # C,<span class="Apple-converted-space"> </span>and C # A < C # B } , in the case all the disjoint unions are well-defined.</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
This is true due to the nature of union, in the case that all the disjoint unions are well-defined.<span class="Apple-converted-space"> </span></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<i><br /></i></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>Symmetry 2:<span class="Apple-converted-space"> </span>(A # B) # C = A # ( B # C) , where either both sides are well-defined or neither are.</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<i><br /></i></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>Symmetry 3: (A x T) # (B x T) = (A # B) x T, where either both sides are well-defined or neither are</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<i><br /></i></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>Symmetry 4: (A x B) x C = A x (B x C)</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Finally, consider an ordered chain of individuals e.g. A < B < C < T, and use the notation e.g. [A,T] to signify that A precedes T in this chain. <span class="Apple-converted-space"> </span>We can then define a derived chaining operation that acts on adjacent intervals, so that e.g. [A,B] , [B,C] = [A, C].</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
If we use the notation</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>a = [A,B]</i></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>b<span class="Apple-converted-space"> </span>=<span class="Apple-converted-space"> </span>[B, C]</i></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>c = [C, T]</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
then we have</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>Symmetry 5:<span class="Apple-converted-space"> </span>(a, b), c = a, (b, c)</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
which works unproblematically in our setting, as the distinction between ordinary and disjoint union is not relevant.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
Looking at the mapping between individuals and values in the context of Knuth and Skilling's mapping between lattice elements and numerical values, how can we interpret</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>c = a + b corresponds to C = A # B</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
in the present context? To calculate c = a + b if A and B are known, one would</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
</div>
<ul style="text-align: left;">
<li>find z_a in A so that v(z_a, A) = a</li>
<li>find z_b in B so that v(z_b, B) = b</li>
<li>let z_c = z_a # z_b</li>
</ul>
<br />
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Then, via the coherence rule</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
z_c = maxzr { v(z_C, C) | x in C }</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
and one can set</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>c = v(z_C,C)</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
The treatment of direct product and chain composition in <i>Foundations of Inference</i> carries over directly here, as there is nothing different about direct products and inclusion in our setting versus the setting they consider.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Axioms 1-5 from Foundations of Inference Section 4 appear to follow directly, the only caveat being that the equations are only to be used when the individuals involved are disjoint.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Section 5.1 in Foundations of Inference deals specifically with the case of disjoint arguments, which is the case of core interest here.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
The overall conclusion is: If v is a coherent value function, then value-assignments of the form</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>r(A) = max{ v(x,A)<span class="Apple-converted-space"> </span>| x in A}</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
should behave like monotone scalings of probabilities.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
This means, that, for instance, they should obey the formula</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>r(S1 # … # Sn) = r(S1) + … + r(Sn)</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
-- or else</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>r(S1 # … # Sn) = r(S1) +^ … +^ r(Sn)</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
where</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>a +^ b<span class="Apple-converted-space"> </span>= f( f^{-1}(a) + f^{-1}(b) )</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
for some monotone function f.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<h2 style="text-align: left;">
<b>Using Incoherent Value Systems to Seed Learning of Coherent Value Systems</b></h2>
</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Now let's get to the punchline -- given the above notion and characterization of value coherence, how might one create a coherent value system that still retains some of the core qualitative aspects of an incoherent value system such as, say, current human value systems?</div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Given an incoherent value system v, one can define a related, derived coherent value system v' as follows.<span class="Apple-converted-space"> </span></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
The basic idea is to define an error function E1(r’) via the sum over all pairs (S1, S2) of</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>( r’(S1 # S2) - ( r’(S1) + r’(S2) ) )^2</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
and another error function E2(q) as the sum over all S of</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>( w(S) * ( r’(S) - r(S) )<span class="Apple-converted-space"> </span>)^2</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
where w(S) is an a priori weight specifying how much a given individual S is valued — e.g. S could be valued proportional to simplicity or proportional to relatedness to a specific base system, etc. ...</div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
[this could be made more sophisticated, e.g. via accounting for intersection of different S in various ways, but this simple version will be sufficient for making the current conceptual points].</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
… and then look for Pareto optima of the problem of minimizing E1 and E2.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
One can then use an iterative algorithm to find a v’(x,A) leading to r’(S) that live on this Pareto frontier, using the original v(x,A) as an initial condition.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
This is somewhat analogous to Eliezer Yudkowsky’s notion of “coherent extrapolated volition”, but much more clearly defined.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
The optimal iterative algorithm to use here is not clear and this is likely a quite subtle question as the intersection of machine learning/reasoning and numerical analysis. <span class="Apple-converted-space"> </span>However, some simple evocative thoughts pointing in the direction of an appropriate heuristic algorithm may be conceptually interesting.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Along these lines, one can think about an iterative algorithm of the following nature.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Given A and A = B#C, let</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>v1(A | B)</i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
denote the maximum value for A that is achievable via choosing the maximum-value instance of B, and then choosing the maximal-value instance of C that can co-exist with this.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Qualitatively, this means: How much value can we provide for A via maximizing the value of some sub-individual B of A.<span class="Apple-converted-space"> </span>For instance, how much value can we provide for me by first maximizing value for my lungs, or how much value can we provide for my family by first maximizing value for me personally?</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Given E = A # D, let</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>v2(A | E)<span class="Apple-converted-space"> </span></i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
denote the maximum value for A that is achievable via choosing the maximum-value instance of E, and then choosing the maximal-value instance of D that can co-exist with this, and let</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<i>v3(A | D)<span class="Apple-converted-space"> </span></i></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
denote the maximum value for A that is achievable via choosing the maximum-value instance of D, and then choosing the maximal-value instance of E that can co-exist with this.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
These measure: How much value can we provide for A via maximizing the value of some individual containing A, or of some individual that is composed with A to form a commonly containing individual?<span class="Apple-converted-space"> </span>For instance, how much value can we provide for me via first maximizing the value of my family, or via first maximizing the value of the other people in my family?</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
If v is coherent, then v1=v2=v3=v.<span class="Apple-converted-space"> </span></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
In general, one could think about using (v1 + v2 + v3) (A) as an estimator for v’(A) to help guide the iterative optimization algorithm. <span class="Apple-converted-space"> </span>This sum (v1 + v2 + v3) (A)<span class="Apple-converted-space"> </span>is an estimate of the value providable for A via maximizing the value of a randomly chosen sub-individual, super-individual or connected-individual of A. <span class="Apple-converted-space"> </span>This will often be a useful pointer in the direction of a more coherent value system than v, i.e. (v1 + v2 + v3) () is likely to be more coherent than v().</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
This particular estimator is relatively crude and much more sophisticated, qualitatively similar estimators can surely be created.<span class="Apple-converted-space"> </span>But the idea I want to get across is that iterative pursuit of a coherent value system that is close to a given incoherent value system, with search seeded from this incoherent value system, may involve iterative steps through intermediate value systems that are conceptually reminiscent of the thinking behind CEV.<span class="Apple-converted-space"> </span>That is, one can think about</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
</div>
<div style="text-align: left;">
</div>
<div style="text-align: left;">
</div>
<ul style="text-align: left;">
<li>What kind of people would current humans like to be, if they could more fully realize their own values</li>
</ul>
<ul style="text-align: left;">
<li>What would these hypothetical “revised better humans” value, and what kind of humans would THEY like to be</li>
</ul>
<ul style="text-align: left;">
<li>etc.</li>
</ul>
<br />
<br />
<br />
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
This sort of iterative process, while rough and poorly-defined, is similar to v1 + v2 + v3 as defined above, and could be interesting as an avenue for iterating from current incoherent human values to a coherent value system living on the above-described Pareto frontier.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<h2 style="text-align: left;">
<b>Varieties of Coherent Value System</b></h2>
</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Assuming there are multiple coherent value systems on the Pareto frontier, then one could guide the iterative search process toward a coherent value system in various different ways.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
For instance, referencing the above simple estimator for simplicity of discussion, in constructing v1, v2 and v3 one could weight certain A, D and E more highly.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
If one considers this weighting to be achieved via some cost function c(A) applied to individuals A, then one can think about the way different choices for c may impact the ultimate coherent value system obtained.<span class="Apple-converted-space"> </span>(Of course the weight c() could be chosen the same as the weight w() used in the error function itself, and this would probably be the optimal choice in terms of effective guidance of optimization, but it’s not the only choice.)</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
E.g. a “selfish” v’ could be obtained by using a c that weights those S relating to a specific person very highly, and other S much less. <span class="Apple-converted-space"> </span></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
A consistently short-term-gratification oriented v’ could be obtained by using a c that weights S restricted to short periods of time very highly, and S restricted to longer periods of time much less.<span class="Apple-converted-space"> </span>In many cultures this would rule out, e.g. a value system that values being happily married over the time-scale of years, but over the time-scale of hours values having sex with whomever one finds attractive. <span class="Apple-converted-space"> </span>But a purely hedonistic value system that values a long period of time precisely according to the sum of the time-localized pleasures experienced during that period of time, may be perfectly coherent. <span class="Apple-converted-space"> </span>Just as there can be a coherent value system that puts value on a time-local experience based substantially on both its immediate characteristics and its contribution to longer-term goals.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
A value system that puts extremely high value on freedom of choice for individuals, but also extremely high value on societal order and structure, may be incoherent within the scope of human societies feasible in the context of modern human psychology and culture. <span class="Apple-converted-space"> </span>A value system that prioritizes order and structure for society and obedience and submission for individuals is more likely to be coherent, as is one that values both freedom of choice and creative anarchic social chaos. <span class="Apple-converted-space"> </span>The professed value systems of most contemporary influential political parties are, in this sense, obviously extremely incoherent.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
<h2 style="text-align: left;">
<b>Are Intelligences with Coherent Value Systems More Efficien</b>t?</h2>
</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
Arguably an intelligent system that directs its actions according to a coherent value system will, all else equal, be more efficient than one that directs its actions according to an incoherent one.<span class="Apple-converted-space"> </span>This is because a mind with an incoherent value system will choose actions oriented toward maximizing value of one subset S1 of the world, and then later choose actions oriented toward maximizing some other subset S2 of the world — and will find that what it did in the context of S2 acts against what it did in the context of S1, and vice versa.<span class="Apple-converted-space"> </span>Whereas for a mind with a coherent value system, actions chosen with respect to different subsets of the world will tend to reinforce and support each other, except where inference errors or unexpected properties of the world intervene.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
This argument suggests that, in an evolutionary context involving competition between multiple intelligences, there will be a certain advantage to the ones with coherent value systems. <span class="Apple-converted-space"> </span>However, this advantage doesn’t have to be decisive, because there may be other advantages enjoyed by entities with incoherent value systems. <span class="Apple-converted-space"> </span>For instance, maintaining a coherent value system may sometimes be highly expensive in terms of space, time and energetic resources (it can be quite complex to figure out the implications of one’s actions for all the subsets of the world they impinge upon!).<span class="Apple-converted-space"> </span></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p3" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal;">
My suspicion is that as computational and energetic resources become more ample and easily accessible by the competing cognizers in an evolutionary system, the efficiency advantage of a coherent value system becomes an increasingly significant factor. <span class="Apple-converted-space"> </span>This suspicion seems a very natural and important candidate for further formal and qualitative exploration.</div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
<div class="p4" style="-webkit-text-stroke-color: rgb(0, 0, 233); font-family: Helvetica; font-size: 12px; font-stretch: normal; font-variant-east-asian: normal; font-variant-numeric: normal; line-height: normal; min-height: 14px;">
<br /></div>
</div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com5tag:blogger.com,1999:blog-11168555.post-1405760873751809072020-06-16T22:49:00.001-04:002020-06-16T22:51:42.359-04:00Simplificational Causal Decision Theory <div dir="ltr" style="text-align: left;" trbidi="on">
In a few spare moments lately, I found myself revisiting issues regarding the foundations of decision theory, and came up with a few ideas that are somewhat new and maybe useful.<br />
<br />
TL;DR version is:<br />
<br />
<br />
<ul style="text-align: left;">
<li>I define "simplificational causality", as a slight generalization of the definition of causality using algorithmic Markov conditions</li>
<li>I give a (semi-)formalization that tells a system <b><i>what properties it could potentially possess that would be (simplificationally) causal for the situation of being embedded in desirable possible universes.</i></b></li>
</ul>
<br />
<br />
I think this is as close as one can come to a meaningful reasonably-general-purpose decision theory without getting boggled up with delusions of free will and such.<br />
<h2 style="text-align: left;">
Boring Historical Prelude</h2>
<br />
The last time I plunged into decision theory issues was a decade ago when I wrote a draft called "<a href="https://goertzel.org/CounterfactualReprogrammingDecisionTheory.pdf">Counterfactual Reprogramming Decision Theory</a>" (CRDT) which I was somewhat but not entirely happy with. The abstract of that paper read:<br />
<br />
<i>"A novel variant of decision theory is presented. The basic idea is that one should ask, at each point in
time: What would I do if the reprogrammable parts of my brain were reprogrammed by a superintelligent
Master Programmer with the goal of supplying me with a program that would maximize my utility averaged
over possible worlds? Problems such as the Prisoner’s Dilemma, the value of voting, Newcomb’s Problem
and the Psychopath Button are reviewed from this perspective and shown to be addressed in a satisfactory
way."</i><br />
<br />
<i>(That plunge into decision theory was largely triggered by the legendary transhumanist John Oh who, after the first AGI conference in 2006, hounded me relentlessly for a solution to voting paradoxes and other such decision-theory conundrums!)</i><br />
<br />
At some point in the subsequent years I decided that CRDT was basically the same as UDT2, an improved version of Wei Dai's Updateless Decision Theory. For background see<br />
<br />
<ul style="text-align: left;">
<li><a href="https://wiki.lesswrong.com/wiki/Updateless_decision_theory">https://wiki.lesswrong.com/wiki/Updateless_decision_theory</a></li>
<li><a href="https://www.lesswrong.com/posts/zztyZ4SKy7suZBpbk/another-attempt-to-explain-udt">https://www.lesswrong.com/posts/zztyZ4SKy7suZBpbk/another-attempt-to-explain-udt</a></li>
<li><a href="https://www.lesswrong.com/posts/zd2DrbHApWypJD2Rz/udt2-and-against-ud-assa">https://www.lesswrong.com/posts/zd2DrbHApWypJD2Rz/udt2-and-against-ud-assa</a></li>
</ul>
<br />
UDT and UDT2 are described in the above links as follows (just to give you a flavor if you're too lazy/busy to click on the links)<br />
<br />
<i>"More formally, [in UDT] you have an initial distribution of "weights" on possible universes (in the currently most general case it's the Solomonoff prior) that you never update at all. In each individual universe you have a utility function over what happens. When you're faced with a decision, you find all copies of you in the entire "multiverse" that are faced with the same decision ("information set"), and choose the decision that logically implies the maximum sum of resulting utilities weighted by universe-weight. </i><br />
<i><br /></i>
<i>"UDT1 receives an observation X and then looks for provable facts of the form "if all my instances receiving observation X choose to take a certain action, I'll get a certain utility".</i><br />
<i><br /></i>
<i>"UDT1.1 also receives an observation X, but handles it differently. It looks for provable facts of the form "if all my instances receiving various observations choose to use a certain mapping from observations to actions, I'll get a certain utility". Then it looks up the action corresponding to X in the mapping."</i><br />
<i><br /></i>
<i>... "[W]hile UDT1 optimizes over possible outputs to its input and UDT1.1 optimizes over possible input/output mappings it could implement, UDT2 simultaneously optimizes over possible programs to self-modify into and the amount of time (in computation steps) to spend before self-modification."</i><br />
<br />
<br />
This is weird-ass convoluted stuff but you actually do have to go to these sorts of lengths to avoid various pathologies and paradoxes that emerge from more straightforward "textbook" approaches like evidential and causal decision theory.<br />
<br />
As neither CRDT nor UDT2 was ever fully formalized or fleshed out, it's hard to say (without doing a lot more work) exactly how close to equivalent they are or what the key differences are.<br />
<br />
Anyway now -- with the benefit of a decade of reflection on the fundamental nature of the universe and how to formalize it in simple ways -- and also with the benefit of the concept of algorithmic Markov conditions, due to Janzing and Scholkopf -- I have been taking a different and (I currently think) more fundamental direction regarding these topics.<br />
<br />
<h2 style="text-align: left;">
(Relational) Simplificational Causality</h2>
So, getting the point --<br />
<br />
-- or, firstly, the setup needed to properly articulate the point --<br />
<br />
Consider an ensemble of possible universes, and a certain predicate F which applies to systems within universes (a “system” being, at first approximation, simply a subset of a universe — though there may emerge some reason to restrict this).<br />
<br />
The predicate F defines an “individual”, e.g. Ben Goertzel or the USA, in a way that spans the instances of this individual across multiple universes. If F(S) is true then S is an instance of the individual.<br />
<br />
Assume a meta-observer, which is another predicate G that applies to pairs of the form (system, property), and makes estimates G(S,f) of the degree to which system S displays property f.<br />
<br />
Assume a value function v(F,U), which rates the quality of a universe from the perspective of an individual, so that vv=(F,U) can be inferred by the meta-observer from looking at the state of some instance of F.<br />
<br />
Then the desirability of a universe according to F, should be measured as the value measured by v(F,U) for the instances of F in universe U.<br />
<br />
A property of an individual is characterized by a predicate that assesses the degree to which an instance possesses that property<br />
<br />
The desirability of a property of instances of F, should be measured as the average over universes of:<br />
<br />
<i>(desirability of the universe) * (degree to which the instances of F in the universe have the property) </i><br />
<br />
Now let’s assume a simplicity measure s(F,A,B), which measures the conditional simplicity of predicate A relative to predicate B, from the perspective of F (meaning that the meta-observer can identify the assessment of s(A|B) relative to an instance of F via inspecting that instance). <br />
<br />
<div style="text-align: left;">
(In "<span style="background-color: white; font-family: inherit;"><a href="https://arxiv.org/abs/2004.05269">Grounding Occam's Razor in a Formal Theory of Simplicity</a>" </span>I gave one approach to defining a set of properties that a function should have to be considered a useful simplicity measure.)</div>
<br />
We can define the mutual simplification of x and y conditional on z as<br />
<br />
<i>I(x:y| z)= s(y|z) = s(x|z) + s(y|z) - s(x,y |z)</i><br />
<br />
And then... ba-da-bing! ... we can construct causal networks, e.g. between properties of individuals, using the postulate that if x and y have nonzero mutual simplification, they must have some common cause.<br />
<br />
This is what I'd call “simplificational causality”.<br />
<br />
(Note that the above is a minor generalization/abstraction of the algorithmic Markov condition approach to causality outlined in <a href="https://is.tuebingen.mpg.de/fileadmin/user_upload/files/publications/paper_IEEE_version3_webseite_6526%5B1%5D.pdf">Causal inference usingthe algorithmic Markov condition</a>... an outstanding paper that I recommend strongly...)<br />
<br />
For a given degree d of desirability, we can then ask: <i>What is the simplest property that instances of F may have, that will be simplificationally causal for these instances to have desirability of degree at least d?</i><br />
<br />
So -- what does all this cockamamie abstraction tell us about decision theory?<br />
<br />
OK: This is not exactly telling an instance of F what decision to make in a given circumstance. <br />
<br />
But it IS telling an instance of F <b><i>what properties it could potentially possess that would be (simplificationally) causal for the situation of being embedded in desirable possible universes.</i></b><br />
<br />
Now I suspect that if you measure simplification using algorithmic information, then when you work out the math of algorithmic Markov conditions, you'll wind up with something in the close vicinity of UDT2 and CRDT. However, even if so (as usual I don't have free time to work the details), I think the formulation I've given here is more conceptually elegant and transparent.<br />
<br />
<h2 style="text-align: left;">
Practical Approximations/ Applications?</h2>
It's not why I started musing in this direction, but I think there may actually be some use for these ideas in the SingularityNET / Rejuve team's currently work using OpenCog for causal network inference in a biological context (soon perhaps to be extended to other contexts such as robotics).<br />
<br />
We are estimating relative simplification in OpenCog now using fuzzy pattern-sets constructed from nodes and links in OpenCog Atomspace (this is part of the PLN logic systems intensional inference). So we have some practical ways to estimate simplification, and in this context we could estimate the simplificational causality between two biological actors, e.g. two proteins playing roles in protein interaction networks, which can be useful e.g. in automated discovery of new biological pathways.<br />
<br />
(Of course this practical work involves simplicity measures that are crude compared to e.g. conditional algorithmic information, but they have the advantage of being feasible to estimate...)</div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com0tag:blogger.com,1999:blog-11168555.post-13934846138965328402020-05-12T20:46:00.002-04:002020-05-13T11:11:29.428-04:00Morphic Anti-Resonance<div dir="ltr" style="text-align: left;" trbidi="on">
<br />
Morphic resonance — in which patterns that have previously occurred are more likely to re-occur — is a powerful force, characteristic of human minds and cultures and also of the quantum world (cf Smolin’s Precedence Principle)<br />
<br />
But it’s also interesting to think about cases in which morphic anti-resonance holds...<br />
<br />
I.e. with morphic anti-resonance, when a pattern occurs, it is then LESS likely than otherwise would have been the case, to occur again...<br />
<br />
Advanced financial markets could perhaps be like this (because a pattern once it's occurred is an exploitable behavior, so whomever sees the pattern has already been exploited may be extra-disinclined to enact it again)<br />
<br />
The decline effect in psi could also be like this... once an experiment has worked, anti-resonance will cause it to stop working…<br />
<br />
Now, it might seem anti-resonance is also a meta-level regularity expoitable by intelligence ... except that via reflexive application to itself, once anti-resonance kicks in, it will kick itself out ;D<br />
<br />
Trickstery indeed…<br />
<br />
What if clusters of morphic resonance are somehow balanced by clusters of morphic anti-resonance, leaving the overall cosmos morphically neutral-on-average but wildly high-variance...?<br />
<br />
<b>Toward a Formal Model of This Madness/Anti-Madness</b><br />
<br />
If we look at the distribution over patterns in the multiverse, where p(R) indicates the probability of observing pattern R during a certain big chunk of spacetime, then<br />
<br />
Compared to a multiverse with no such oddities,<br />
<br />
<br />
<ul style="text-align: left;">
<li>A multiverse w/ morphic resonance will have a more pointy, peaked (i.e. lower entropy) distribution p()</li>
<li>A multiverse w/ morphic anti-resonance will have a flatter (i.e. higher entropy) distribution p()</li>
</ul>
<br />
<br />
So if we assume we have a multi-multiverse described as a probability distribution over multiverses, then we may posit that the average multiverse has a no-resonance p(), but this is achieved via having some multiverses with higher-entropy p() and some with lower-entropy p()<br />
<br />
Path integrals must then be taken in the multi-multiverse not any base multiverse<br />
<br />
Psi would then be a mix of<br />
<br />
<ul style="text-align: left;">
<li>Morphic resonance and anti-resonance phenomena</li>
<li>Shifts from one multiverse to another, which involve shifts in the entropy of the multiversal pattern-probability distribution</li>
</ul>
<br />
Also note -- an observing mind's reality may be a paraconsistent patchwork of probability distributions (multiverses) rather than a single consistent multiverse...<br />
<br />
<b>Dialectics of Creativity</b><br />
<br />
Going further out on the limb -- Perhaps morphic resonance and anti-resonance enact the dialectic dance of creation vs. destruction?<br />
<br />
I am reminded of economics approaches where <a href="https://phys.org/news/2014-04-econophysics-antimoney-financial-crisis.html">anti-money is used in place of debt,</a><br />
<span class="sewu5kp2yr2qw4"><br /></span>
<br />
<div>
In economics the conserved-ish quantity is money, in physics it's energy, and in morphic pattern-omics it's synchronicity (degree of spooky resonance). </div>
<div>
<br /></div>
<div>
Conservation of synchronicity suggests that a bit of morphic resonance over here is balanced out by a bit of morphic anti-resonance over there. That's how morphic resonance can exist without morphically resonating the cosmos into a repetitive mush...</div>
<ul style="text-align: left;">
</ul>
</div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com0tag:blogger.com,1999:blog-11168555.post-48540668494016496122020-05-12T02:48:00.000-04:002020-05-12T02:48:06.979-04:00GTGI -- General Theory of General Intelligence... coming gradually...<div dir="ltr" style="text-align: left;" trbidi="on">
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
In the background, using spare time here and there, over the last few years I've managed to write down a series of sketchy research papers summarizing key aspects of what has been a long-running thread in my mind for a very long time: A general theory of general intelligence. <o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
And by that I mean a REALLY REALLY general theory of general intelligence ... including the phenomenological aspect of "what it is to be a general intelligence" ... including consciousness from first, second and third person perspectives ... and including the dynamics via which minds help construct each other, and minds and physical reality co-create each other. But also encompassing practical information about how human brains achieve general intelligence and why they achieve it the way they do, and how to effectively create general intelligence in various synthetic substrates, such as computer software.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
I certainly don't claim to be there yet. However, after a few sketchy papers hastily typed out in late nights during the last year, I feel like I finally have a complete outline of such a theory. I know what needs to be in there, and quite a lot of what should be rigorous theorems in such a theory, I now have at least in the form of rough but explicitly articulated conjectures.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
In this blog post I'm going to briefly run through these various papers and explain how I believe they build together toward a GTGI. I'll also highlight some of the gaps that I think will need to be filled in to complete the GTGI story along these lines.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
Starting from the philosophical, this paper<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<a href="https://journals.sfu.ca/jnonlocality/index.php/jnonlocality/article/view/65" style="color: purple;">EURYPHYSICS: A (SOMEWHAT) NEW CONCEPTUAL MODEL OF MIND, REALITY AND PSI</a><o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
outlines a high level perspective on "life, the universe and everything" that bridges cognitive science, theoretical physics, analytical philosophy, phenomenological philosophy and more.</div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
In part this paper was intended as a sequel to the book<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<a href="https://www.amazon.com/Evidence-Psi-Thirteen-Empirical-Research/dp/0786478284" style="color: purple;">The Evidence for Psi</a><o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
that I co-edited with Damien Broderick. The book reviews some of the copious evidence that psi phenomena exist, but doesn't try to explain how they might work. The Euryphysics paper tries to outline a world-model within which a rational yet non-reductive explanation of psi might be constructed -- by constructing a very broad world-model going beyond traditional categories such as physical mental and cultural reality.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
"Euryphysics" means "the wider world" -- a core concept is that there is a broader domain of existence of which our 4D spacetime continuum and our individual minds are just small parts. The regularities governing this broader domain are not entirely "physics-like" (evolution described by concise sets of differential equations) and may be more "mind-like" in some sense. Aspects of "consciousness" may be best considerable at the level of euryphysics rather than physics or individual psychology.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
But how to build a theory of the Eurycosm? (Among other things, this could be -- a theory explaining interesting details about how mind and physical reality create each other.)<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
Let's start with elementary observations. The most elementary sort of observation is a distinction -- just an act of distinguishing some stuff from some other stuff. (Yes, I had some influence from G. Spencer Brown and his friend Lou Kauffmann.) This paper<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<a href="https://arxiv.org/abs/1902.00741" style="color: purple;">Distinction Graphs and Graphtropy: A Formalized Phenomenological Layer Underlying Classical and Quantum Entropy, Observational Semantics and Cognitive Computation</a><o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
introduces a theory of "distinction graphs" -- in which a link is drawn between two observations, relative to a given observer, the observer cannot distinguish them (while it's not clarified in the paper, basically an "observation" can be considered as "something that can be distinguished"). Graphtropy is introduced as an extension of logical entropy from partitions to distinction graphs, along with extensions like probabilistic and quantum distinction graphs. An analogue of the maximum entropy principle for distinction graphs is suggested.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
Graphtropy gives a broad view of surprisingness, which has many values including giving a foundation for notions of time. As suggested in the Euryphysics paper, local time flow may be interpreted in terms of surprisingness gradients, and global time axes via stitching together compatible local time flows.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
Among the various tools that can be useful for analyzing distinction graphs and associated structures is the notion of simplicity vs. complexity. This paper<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<a href="https://arxiv.org/abs/2004.05269" style="color: purple;">Grounding Occam's Razor in a Formal Theory of Simplicity</a><o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
attempts an axiomatic characterization of "what is simplicity"? Conventional simplicity measures like minimal program length emerge as a special case.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
Simplicity allows articulation of "what is a pattern"? (A pattern is a representation-as-something-simpler.)<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
And this allows a nice formalization of the Peircean idea of "the tendency to take habits" -- which is equivalent to Smolin's Precedence Principle in quantum mechanics, or Sheldrake's morphic resonance principle, a plausible high level explanation for psi phenomena.</div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
One would also like to construct something like probability theory that is natural on graphs (e.g. distinction graphs), in the same way that conventional probability is natural on sets. In this paper (inspired heavily by Knuth and Skilling's classic paper Foundations of Inference and its sequels),<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<a href="https://arxiv.org/abs/1703.04382" style="color: purple;">Cost-Based Intuitionist Probabilities on Spaces of Graphs, Hypergraphs and Theorems</a><o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
I bite this bullet, giving a specific way of constructing intuitionistic "generalized probabilities" on top of graphs, hypergraphs, metagraphs and similar structures. The approach relies on some way of assigning "costs" to different graph transformations -- which is provided e.g. if one has a simplicity measure in hand.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
It's also the case that if the nice symmetries needed to construct probabilities only hold approximately for a given domain -- then you get an uncertainty measure on that domain that is approximately probabilistic. I.e. the dependence of probability theory's rules on the underlying symmetry axioms is reasonably smooth, as I argued here:<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-family: Times; font-size: 10pt;"><a href="https://www6.inrae.fr/mia-paris/content/download/3766/36508/version/1/file/Goertzel.pdf" style="color: purple;">Probability Theory Ensues from Assumptions of Approximate Consistency: A Simple Derivation and its Implications for AGI</a><o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-family: Times; font-size: 10pt;">(I only explicitly considered the case of classical probability theory, but the same arguments would hold for the intuitionistic case.)<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-family: Times; font-size: 10pt;">Once you have probabilities, you have second order, third order and then... infinite-order probabilities (defined as distributions over spaces of infinite-order probabilities):<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-family: Times; font-size: 10pt;"><a href="https://goertzel.org/papers/InfiniteOrderProbabilities.pdf" style="color: purple;">Modeling Uncertain Self-Referential Semantics with Infinite-Order Probabilities</a><o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
Are these useful? Well one can construct interesting models of aspects of phenomenological experience, using non-well-founded set theory (aka hypersets),<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<a href="https://www.researchgate.net/publication/267418581_Hyperset_Models_of_Self_Will_and_Reflective_Consciousness" style="color: purple;">Hyperset Models of Self, Will and Reflective Consciousness</a><o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
and layering uncertainty onto these models, you get infinite-order probabilities.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
There is some unification not yet written out here: The hypersets I consider are modeled by apg's ("accessible pointed graphs", i.e. a digraph with a distinguished node N from which all other nodes can be reached), and a directed distinction graph can be interpreted as patchwork of apg's. One can build apg's up from distinction graphs, though I haven't written up that paper yet. Basically you model distinctioning as a directional process -- you ask if an observer already has made observation A, is it able to make observation B considering B as distinct from A? This gives a directed distinction graph, which is then a patchwork of apg's, i.e. a mesh of overlapping hypersets.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
Given probability distributions and simplicity measures, one can start measuring intelligence in traditional ways ("traditional" in the sense of <a href="http://www.vetta.org/documents/Machine_Super_Intelligence.pdf" style="color: purple;">Legg and Hutter</a> or my first book <i>The Structure of Intelligence</i>) ... one can look at intelligence as the ability to achieve complex goals in complex environments using limited resources...<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<a href="http://citeseerx.ist.psu.edu/viewdoc/download?rep=rep1&type=pdf&doi=10.1.1.212.6422" style="color: purple;">Toward a Formal Characterization of Real-World General Intelligence</a><o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
Though it is also worth keeping in mind the wider nature of intelligence as Weaver articulated so richly in his PhD thesis<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<a href="https://www.academia.edu/36140514/Open-Ended_Intelligence_-_PhD_Thesis" style="color: purple;">Open-Ended Intelligence</a><o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
Another paper I haven't yet written up is a formalization of open-ended intelligence in terms of richness of pattern creation. <o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
One can formalize the three key values of "Joy, Growth and Choice" in terms of graphtropy and pattern theory (Joy is patterns continuing, growth is new pattern being created, choice is graphtropy across pattern space) -- so relative to any local time-axis one can look at the amount of Joy/Growth/Choice being manifested which is one way of looking at the amount of open-ended intelligence.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
One way to move from these intriguing generalities toward specific cognitive, computational and physics theories is to assume a specific computational model. In this paper<o:p></o:p></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<a href="https://arxiv.org/abs/2004.05268" style="color: purple;">Combinatorial Decision Dags: A Natural Computational Model for General Intelligence</a><span style="font-family: Times; font-size: 10pt;"><o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-family: Times; font-size: 10pt;">I articulate what seems an especially natural computational model for general intelligence (CoDDs, Combinatorial Decision Directed-acyclic-graphs), and I conjecture that if one assumes this computational model, then some nice compatibilities between graphtropic measures of complexity and simplicity-theoretic measures of complexity emerge. (Actually the paper talks about correlating algorithmic information with logical entropy but the generalization to graphtropy is not a big leap.)<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-family: Times; font-size: 10pt;">A CoDD is basically a decision tree that is recursively nested so that a whole decision tree can serve as an input to a decision tree, and augmented with the the ability to replace two identical subtrees with two instances of a certain token (memo-ization). Repetition-replacement and recursion are enough to tweak decision trees into a Turing-complete computational model (which is basically the insight that SK-combinator calculus is universal, phrased a bit differently).<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-family: Times; font-size: 10pt;">This computational model also leads to some interesting extensions of the basic model of pattern as "representation as something simpler", including the notion of "quattern" -- the quantum analogue of a classical pattern.<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-family: Times; font-size: 10pt;">The paper doesn't draw any connections with distinction graphs -- but it's quite interesting to look at CoDDs whose leaves are observations related in a distinction graph.<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-family: Times; font-size: 10pt;">My primary focus is on applying these GTGI-ish ideas to AI and cognitive science, but the applications to physics also can't be overlooked. In this verrrry sketchy notes-to-self type paper<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<a href="https://goertzel.org/papers/goertzel_information_geom_physics_v3.pdf" style="color: purple; font-family: Times; font-size: 10pt;">Physics as Information Geometry on Causal Webs</a></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-family: Times; font-size: 10pt;">I outline a possible path to creating unified (standard model + gravity) physics models via hypergraph models (including hypergraph links with causal interpretation). Spacetime is a hypergraph and event probabilities are estimated using Feynman type sums that add up terms corresponding to multiple spacetimes as well as multiple possible scenarios within each spacetime. <o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-family: Times; font-size: 10pt;">Ben Dribus, a mathematician who has developed his own much more in-depth <a href="https://www.springer.com/gp/book/9783319500812" style="color: purple;">graph-based physics models</a>, has (in a personal communication) sketched a dynamical equation that works in my causal web model.<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-family: Times; font-size: 10pt;">Another paper not yet written up regards the formal similarities between conservation of energy in physics and conservation of evidence (i.e. avoidance of double counting of evidence) in logic. One can view energy as the form that observation takes in a certain logic (that has observational semantics), and then physical dynamics as a process of derivation in this logic, with the consistency of the logic depending on the conservation of energy (which avoids double-counting evidence).<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-family: Times; font-size: 10pt;">Extending this physics-ish line of thinking in a direction that also encompasses the cognitive, was a recent paper with a messy title:<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<a href="https://arxiv.org/abs/2005.04589" style="color: purple;">Maximal Algorithmic Caliber and Algorithmic Causal Network Inference: General Principles of Real-World General Intelligence? </a><o:p></o:p></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-size: 10pt;">The basic idea here was to come up with physics-ish "dynamical laws of cognition" by replacing Shannon algorithm in MaxEnt type principles, with algorithmic information. Not yet done is to extend this to graphtropy -- by extending Maximum Caliber Principle to distinction graphs that evolve over time, and then creating a corresponding form of Maximal Algorithmic Caliber that works with Combinatorial Decision Dags whose primitives are observations in a distinction graph.<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-size: 10pt;">The "maximum caliber principle" is extended to a "maximum algorithmic caliber principle" that characterizes the possible worlds most likely to accord with a given set of observations -- one should assume the world has evolved with the maximum algorithmic caliber consistent with observations (basically, the most computationally dense way consistent with observations). Basically, this just means that if you don't know how the world has made your observations come about, you need to make some assumption. Lacking some simplicity prior, there are more possible worlds involving a lot of distinctions than a few, so the odds will be high (based on simple Principle of Indifference type symmetry arguments) that the underlying reality makes a lot of distinctions. Given a simplicity prior, the most likely worlds will be the ones that make about as many distinctions as the prior considers in the "reasonably likely" range.<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-size: 10pt;">Algorithmic Markov processes, the algorithmic-information analogue of ordinary statistical Markov processes, turn out to be the most rational hypothesis to use when inferring processes based on data. There are more possible processes similar to an algorithmic Markov process that obey your given constraints, than any other sort of processes. If you looked in the mind of a near maximally generally intelligent AIXI-tl type agent, you would see that it was implicitly or explicitly making the assumption that the world is often roughly an algorithmic Markov process.<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-size: 10pt;">To move from these highly general "laws of mind" toward laws of human-like mind one needs to look at the special situations for which human-like minds evolved. In the paper<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-size: 10pt;"><a href="https://www.semanticscholar.org/paper/A-Mind-World-Correspondence-Principle-Goertzel/ec4ca4401f38ec10d8d3cf40c39af0a4bfe86557" style="color: purple;">A Mind-World Correspondence Principle</a><o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-size: 10pt;">I suggest that symmetries and other regularities in the environments and goals that an intelligence needs to deal with, should be mappable via (uncertain) morphisms into corresponding symmetries/regularities in the structure and dynamics of the intelligent system itself. I roughly formalize this correspondence in terms of category theory (which ultimately needs an intuitionistic probability-like quantity like the one I mentioned above, which however I only discovered/invented a few years after writing the Mind-World Correspondence paper).<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-size: 10pt;">As for what are the symmetries and regularities human-like minds in particular need to deal with, I made some concrete suggestions in<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-family: Times; font-size: 10pt;"><a href="https://goertzel.org/dynapsyc/2009/EmbodiedCommunicationPrior.pdf" style="color: purple;">THE EMBODIED COMMUNICATION PRIOR: A CHARACTERIZATION OF GENERAL INTELLIGENCE IN THE CONTEXT OF EMBODIED SOCIAL INTERACTION</a></span><span style="font-size: 10pt;"><o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
It should be noted that my suggestions are far more specific than what the great Yoshua Bengio proposed in his "<a href="https://arxiv.org/abs/1709.08568" style="color: purple;">consciousness prior</a>" paper. Basically there he suggests that AGI needs a prior distribution that favors joint distributions that factor into forms where most weight goes to a small number of factors. This is a very sensible idea and does indeed tie in with the way working memory works in current human and AI minds. However, I think the structure and dynamics of human-like minds have been adapted heavily to considerably more specialized assumptions to do with modeling events in 4D spacetime, and specifically to handling communication among spatiotemporally embodied agents who share the same sensation and action space.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
One feature of the environments and goals human-like minds are faced with, is that they tend to factorize into qualitatively different types of knowledge / perception / action -- e.g. procedural vs. declarative/semantic vs. attentional vs. sensory, etc. This leads to minds that have distinct yet closely coupled subcomponents that need to have robust capability to help each other out of difficult cognitive spots -- "Cognitive Synergy", which underpins the<a href="http://opencog.org/" style="color: purple;"> OpenCog</a> AGI design I've been working on for 1-2 decades (depending how you count). The different types of human memory correspond closely to different aspects of the everyday human physical and social environment.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
The Embodied Communication Prior includes "tendency to take habits" as a corollary. This leads to the amusing notion that, via reflexive application of morphic resonance to itself, the human sphere within our physical spacetime may have some "spooky synchronistic correlation" with other portions of the Eurycosm that also happen to display the tendency to take habits!</div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
More prosaically, the paper<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<a href="https://arxiv.org/abs/1703.04361" style="color: purple;">Toward a Formal Model of Cognitive Synergy</a><o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
formalizes the concept of cognitive synergy on a category-theoretic foundation.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
What is not articulated fully there is that, ultimately, the cognitive processing of real-world AGI systems can be viewed as: a set of interacting cognitive algorithms, each of which in a sense results from doing program specialization on the universal algorithm <i>"form an algorithmic Markov model consistent with one's observations, and use it to drive inference about what procedures will achieve one's goals given the observed context",</i> relative to focus on a specific sort of knowledge, memory or situation (e.g. procedural, sensory, declarative...). These specialized cognitive algorithms must be learned/evolved based on multiple constraints including energetic usage, minimizing spatial extent and maximizing processing speed, and interoperability among the different cognitive algorithms (so that they can see each others' internal states so as to help each other out when they get stuck).<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
Design of a framework like OpenCog may be viewed as performing this sort of program specialization "by hand", as we don't have automated program specializers capable of this degree of complexity. An AGI program specializer will be able to do it, but then we have a chicken-egg problem -- which is solved by human AGI system designers performing the first round of the iteration.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
The paper<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<a href="https://arxiv.org/abs/1703.04368" style="color: purple;">Symbol Grounding via Chaining of Morphisms</a><o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
explains how the connection between language, action, perception and memory works in terms of the category-theoretic model of cognitive synergy.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
The paper<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<a href="https://goertzel.org/Neural_Foundations_Symbolic_Thought.pdf" style="color: purple;">How Might the Brain Represent Complex Symbolic Knowledge?</a><o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
gives some speculative ideas regarding how the human brain may implement some of these abstract structures (using multiple neural-net modules interconnected, e.g. different, closely cooperating architectures for corpus and hippocampus -- but not as simplistically interconnected as in currently popular deep or shallow neural net architectures).<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
This lets us revisit the vexed issue of "consciousness." My view is that consciousness is a universal property immanent in all existence, but that "human-like consciousness" has some special properties, which come out of the Embodied Communication Prior along with other factors. This paper<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<a href="https://www.sciencedirect.com/science/article/pii/S1877050914015439" style="color: purple;">Characterizing Human-like Consciousness: An Integrative Approach</a><o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
aims to identify what is special about human-like consciousness as opposed to other flavors. </div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
This includes physical and computationally-cognitive correlates of the hyperset models of self, will and awareness alluded to earlier. Mapping between distinction graphs and hyperset apg's, can be seen as mapping between sensate-oriented and reflection-oriented reflexive meta-views of the same base subjective experience.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
This paper<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-family: Times; font-size: 10pt;"><a href="https://goertzel.org/Goertzel_IJMC_Special_Issue.pdf" style="color: purple;">WHEN SHOULD TWO MINDS BE CONSIDERED VERSIONS OF ONE ANOTHER?</a><o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
deals with the question of identity under conditions of gradual change -- arguing that if a mind changes slowly enough that, at each stage, it models where it came from, where it is and where it's going in terms of a unified self-construct.... then in essence it IS a unified self. This IMO solves the issue of "continuity of consciousness and identity" in a mind uploading context.<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
To realize these abstract GTGI ideas in practical AGI systems, one needs a series of bridging formalisms, toolkits and systems. This is something I'm currently working on within the TrueAGI / Atomese 2.0 research initiative (still an early-stage non-public thing), but one paper has recently crawled out of this particular research swamp:<o:p></o:p></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<br /></div>
<div style="font-size: 10pt; margin-bottom: 0.0001pt; margin-left: 0in; margin-right: 0in;">
<a href="https://arxiv.org/abs/2004.05267" style="color: purple;">What Kind of Programming Language Best Suits Integrative AGI?</a><o:p></o:p></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-size: 10pt;">Among other things, what is advocated there is a gradually typed approach to AI programming, wherein different cognitive processes corresponding to different types of memory/knowledge are realized using different type systems. Casting between these type systems is part of the process of cognitive synergy. <o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-size: 10pt;">There is a Curry-Howard correspondence between a gradually typed language like this, and a paraconsistent logic. As cognitive processes must be probabilistic, what we ultimately have is a Curry-Howard correspondence between intuitionistically-probabilistic paraconsistent logic and a gradually typed probabilistic functional programming language. <o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-size: 10pt;">The intuitionistic aspect of this logic, maps into the absence of highly general continuation-passing features in the language -- and it means that ultimately the logic can be reduced to operations on distinction graphs, and the corresponding programs can be reduced to e.g. CoDDs operating on elementary observations drawn from distinction graphs.<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-size: 10pt;">An AGI-oriented hypergraph knowledge store like the OpenCog Atomspace can be viewed as a CoDD that operates on the elementary observations made by a specific cognitive system, and abstracts from these observations to form programs for generating sets of observations from more compact descriptions. These include observations of what action-combinations tend to lead to what goals in what contexts. A programming language like Atomese 2.0 is a concise, workable way of creating higher level program constructs equivalent ultimately to CoDDs over distinction graphs.<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<span style="font-size: 10pt;">So there you go. Turning all the above papers into a single coherent narrative would be many months of full-time work -- and then turning all the conjectures in the papers into actual theorems would be probably several years of full-time work. I'm not sure when I'll get to all that, since I have an insane number of other things on my plate. But I do feel like it's finally time for the "weaving together and rigorizing" phase of my GTGI quest -- I think that with the most recent few papers, among the ones listed above, the core ideas needed have finally fallen into place!<o:p></o:p></span></div>
<div class="MsoNormal" style="font-family: Cambria; font-size: 12pt; margin: 0in 0in 0.0001pt;">
<br /></div>
</div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com22tag:blogger.com,1999:blog-11168555.post-80157897096811954792020-04-11T12:51:00.000-04:002020-04-11T12:51:12.446-04:00The Likely Nasty Social, Economic and Surveillance Aftereffects of COVID-19 -- and How to Combat Them<div dir="ltr" style="text-align: left;" trbidi="on">
<!--[if gte mso 9]><xml>
<o:OfficeDocumentSettings>
<o:AllowPNG/>
</o:OfficeDocumentSettings>
</xml><![endif]-->
<!--[if gte mso 9]><xml>
<w:WordDocument>
<w:View>Normal</w:View>
<w:Zoom>0</w:Zoom>
<w:TrackMoves/>
<w:TrackFormatting/>
<w:PunctuationKerning/>
<w:ValidateAgainstSchemas/>
<w:SaveIfXMLInvalid>false</w:SaveIfXMLInvalid>
<w:IgnoreMixedContent>false</w:IgnoreMixedContent>
<w:AlwaysShowPlaceholderText>false</w:AlwaysShowPlaceholderText>
<w:DoNotPromoteQF/>
<w:LidThemeOther>EN-US</w:LidThemeOther>
<w:LidThemeAsian>JA</w:LidThemeAsian>
<w:LidThemeComplexScript>X-NONE</w:LidThemeComplexScript>
<w:Compatibility>
<w:BreakWrappedTables/>
<w:SnapToGridInCell/>
<w:WrapTextWithPunct/>
<w:UseAsianBreakRules/>
<w:DontGrowAutofit/>
<w:SplitPgBreakAndParaMark/>
<w:EnableOpenTypeKerning/>
<w:DontFlipMirrorIndents/>
<w:OverrideTableStyleHps/>
<w:UseFELayout/>
</w:Compatibility>
<m:mathPr>
<m:mathFont m:val="Cambria Math"/>
<m:brkBin m:val="before"/>
<m:brkBinSub m:val="--"/>
<m:smallFrac m:val="off"/>
<m:dispDef/>
<m:lMargin m:val="0"/>
<m:rMargin m:val="0"/>
<m:defJc m:val="centerGroup"/>
<m:wrapIndent m:val="1440"/>
<m:intLim m:val="subSup"/>
<m:naryLim m:val="undOvr"/>
</m:mathPr></w:WordDocument>
</xml><![endif]--><!--[if gte mso 9]><xml>
<w:LatentStyles DefLockedState="false" DefUnhideWhenUsed="true"
DefSemiHidden="true" DefQFormat="false" DefPriority="99"
LatentStyleCount="276">
<w:LsdException Locked="false" Priority="0" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Normal"/>
<w:LsdException Locked="false" Priority="9" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="heading 1"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 2"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 3"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 4"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 5"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 6"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 7"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 8"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 9"/>
<w:LsdException Locked="false" Priority="39" Name="toc 1"/>
<w:LsdException Locked="false" Priority="39" Name="toc 2"/>
<w:LsdException Locked="false" Priority="39" Name="toc 3"/>
<w:LsdException Locked="false" Priority="39" Name="toc 4"/>
<w:LsdException Locked="false" Priority="39" Name="toc 5"/>
<w:LsdException Locked="false" Priority="39" Name="toc 6"/>
<w:LsdException Locked="false" Priority="39" Name="toc 7"/>
<w:LsdException Locked="false" Priority="39" Name="toc 8"/>
<w:LsdException Locked="false" Priority="39" Name="toc 9"/>
<w:LsdException Locked="false" Priority="35" QFormat="true" Name="caption"/>
<w:LsdException Locked="false" Priority="10" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Title"/>
<w:LsdException Locked="false" Priority="1" Name="Default Paragraph Font"/>
<w:LsdException Locked="false" Priority="11" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Subtitle"/>
<w:LsdException Locked="false" Priority="22" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Strong"/>
<w:LsdException Locked="false" Priority="20" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Emphasis"/>
<w:LsdException Locked="false" Priority="59" SemiHidden="false"
UnhideWhenUsed="false" Name="Table Grid"/>
<w:LsdException Locked="false" UnhideWhenUsed="false" Name="Placeholder Text"/>
<w:LsdException Locked="false" Priority="1" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="No Spacing"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 1"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 1"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 1"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 1"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 1"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 1"/>
<w:LsdException Locked="false" UnhideWhenUsed="false" Name="Revision"/>
<w:LsdException Locked="false" Priority="34" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="List Paragraph"/>
<w:LsdException Locked="false" Priority="29" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Quote"/>
<w:LsdException Locked="false" Priority="30" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Intense Quote"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 1"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 1"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 1"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 1"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 1"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 1"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 1"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 1"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 2"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 2"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 2"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 2"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 2"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 2"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 2"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 2"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 2"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 2"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 2"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 2"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 2"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 2"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 3"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 3"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 3"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 3"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 3"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 3"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 3"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 3"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 3"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 3"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 3"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 3"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 3"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 3"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 4"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 4"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 4"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 4"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 4"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 4"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 4"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 4"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 4"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 4"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 4"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 4"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 4"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 4"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 5"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 5"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 5"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 5"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 5"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 5"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 5"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 5"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 5"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 5"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 5"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 5"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 5"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 5"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 6"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 6"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 6"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 6"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 6"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 6"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 6"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 6"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 6"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 6"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 6"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 6"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 6"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 6"/>
<w:LsdException Locked="false" Priority="19" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Subtle Emphasis"/>
<w:LsdException Locked="false" Priority="21" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Intense Emphasis"/>
<w:LsdException Locked="false" Priority="31" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Subtle Reference"/>
<w:LsdException Locked="false" Priority="32" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Intense Reference"/>
<w:LsdException Locked="false" Priority="33" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Book Title"/>
<w:LsdException Locked="false" Priority="37" Name="Bibliography"/>
<w:LsdException Locked="false" Priority="39" QFormat="true" Name="TOC Heading"/>
</w:LatentStyles>
</xml><![endif]-->
<style>
<!--
/* Font Definitions */
@font-face
{font-family:"MS 明朝";
mso-font-charset:78;
mso-generic-font-family:auto;
mso-font-pitch:variable;
mso-font-signature:1 134676480 16 0 131072 0;}
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;
mso-font-charset:0;
mso-generic-font-family:auto;
mso-font-pitch:variable;
mso-font-signature:-536870145 1107305727 0 0 415 0;}
@font-face
{font-family:Cambria;
panose-1:2 4 5 3 5 4 6 3 2 4;
mso-font-charset:0;
mso-generic-font-family:auto;
mso-font-pitch:variable;
mso-font-signature:-536870145 1073743103 0 0 415 0;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{mso-style-unhide:no;
mso-style-qformat:yes;
mso-style-parent:"";
margin:0in;
margin-bottom:.0001pt;
mso-pagination:widow-orphan;
font-size:12.0pt;
font-family:Cambria;
mso-ascii-font-family:Cambria;
mso-ascii-theme-font:minor-latin;
mso-fareast-font-family:"MS 明朝";
mso-fareast-theme-font:minor-fareast;
mso-hansi-font-family:Cambria;
mso-hansi-theme-font:minor-latin;
mso-bidi-font-family:"Times New Roman";
mso-bidi-theme-font:minor-bidi;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
mso-themecolor:hyperlink;
text-decoration:underline;
text-underline:single;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-noshow:yes;
mso-style-priority:99;
color:purple;
mso-themecolor:followedhyperlink;
text-decoration:underline;
text-underline:single;}
.MsoChpDefault
{mso-style-type:export-only;
mso-default-props:yes;
font-family:Cambria;
mso-ascii-font-family:Cambria;
mso-ascii-theme-font:minor-latin;
mso-fareast-font-family:"MS 明朝";
mso-fareast-theme-font:minor-fareast;
mso-hansi-font-family:Cambria;
mso-hansi-theme-font:minor-latin;
mso-bidi-font-family:"Times New Roman";
mso-bidi-theme-font:minor-bidi;}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.25in 1.0in 1.25in;
mso-header-margin:.5in;
mso-footer-margin:.5in;
mso-paper-source:0;}
div.WordSection1
{page:WordSection1;}
-->
</style>
<!--[if gte mso 10]>
<style>
/* Style Definitions */
table.MsoNormalTable
{mso-style-name:"Table Normal";
mso-tstyle-rowband-size:0;
mso-tstyle-colband-size:0;
mso-style-noshow:yes;
mso-style-priority:99;
mso-style-parent:"";
mso-padding-alt:0in 5.4pt 0in 5.4pt;
mso-para-margin:0in;
mso-para-margin-bottom:.0001pt;
mso-pagination:widow-orphan;
font-size:12.0pt;
font-family:Cambria;
mso-ascii-font-family:Cambria;
mso-ascii-theme-font:minor-latin;
mso-hansi-font-family:Cambria;
mso-hansi-theme-font:minor-latin;}
</style>
<![endif]-->
<!--StartFragment-->
<br />
<div align="center" class="MsoNormal" style="text-align: center;">
<br /></div>
<div class="MsoNormal">
A lot of attention right now is going into the question of
flattening the curve of global COVID-19 infection -- and this is exactly
right.<span style="mso-spacerun: yes;"> </span>I've been trying to do my own
part here, via organizing the <a href="https://covidathon.devpost.com/">COVIDathon</a>
blockchain-AI-against-COVID-19 hackathon, and working with<span style="mso-spacerun: yes;"> </span>my SingularityNET colleagues on using some of
our AI code for simulating COVID-19 spread and analyzing related biology.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
It's also important, though, to think about the other side
of the curve -- what happens once the virus starts to gradually recede into the
background, and life resumes some variation of "normal."<span style="mso-spacerun: yes;"> </span>How will things be different after
COVID-19?<span style="mso-spacerun: yes;"> </span>Which of the unusual things
happening now in the midst of the pandemic are likely to continue to have
impact in the post-pandemic world?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
TL;DR it seems the answer is: Barring something unusual and
countervailing happening, the impact of the pandemic will be the rich getting
richer, the poor getting poorer, and Big Tech and Big Government getting more
access to diverse personal data and more skill at mining it effectively.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Potentially these effects could be palliated by rolling out
decentralized blockchain-based technologies for managing aspects of the
pandemic and the pandemic-era economy.<span style="mso-spacerun: yes;">
</span>But it appears likely that, even if we succeed in getting a few such
technologies built and adopted rapidly via COVIDathon and other efforts, by and
large it will be centralized technologies, centralized government agencies and
companies and the traditionally financialized economy that will dominate
COVID-19 response.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
A more open question is whether, when the next pandemic or
other global crisis rolls around, decentralized tech will be ready to play a
major role.<span style="mso-spacerun: yes;"> </span>Can COVID-19 and its
impacts on society, economy and industry serve as a wake-up call regarding the
risks global crises pose on multiple fronts including data sovereignty and
economic fairness?<span style="mso-spacerun: yes;"> </span>Will this wake-up
call be loud enough to rouse a large open-source development community into action
regarding the creation of decentralized, secure and democratically controlled
technologies for doing things like, say, managing uploaded personal medical
data ... tracking and predicting spread of epidemics ... carrying out precision
medicine analytics on clinical trials<o:p></o:p></div>
<div class="MsoNormal">
<span style="mso-spacerun: yes;"> </span>...<span style="mso-spacerun: yes;"> </span>and assessing lifestyle choices in the light
of current medical realities and practicalities like weather and
transportation?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Let's run through the probable future in more detail.<span style="mso-spacerun: yes;"> </span>Social distancing and travel restrictions are
most likely to cause the virus's spread to slow as 2020 progresses; and then
before too long effective antiviral compounds or cocktails will be available.<span style="mso-spacerun: yes;"> </span>Sometime in 2021, most likely, COVID-19
vaccines will hit the market; and then this virus currently wreaking so much
havoc will be relegated to a status much like that of the lowly flu.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
In the meantime, though lots of low-wage service workers are
getting laid off ... and many will not get re-hired, as many businesses will
choose to rebuild in other ways after the pandemic fades (automation, anyone?).<span style="mso-spacerun: yes;"> </span>For instance, many of the people who are now
ordering groceries for home delivery for the first time, will continue doing
this a lot after COVID-19 is gone.<span style="mso-spacerun: yes;">
</span>Resulting in fewer jobs for supermarket cashiers and other staff.<span style="mso-spacerun: yes;"> </span>The same sort<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
At the same time, savvy investment funds are right now
buying up every valuable asset they can at bargain prices -- so that after the
pandemic fades they will own an even larger percentage of the planet<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
And the techlash is already fading into the dim recesses of
history along with net neutrality -- as everyone grows increasingly attached to
Amazon, Netflix, Google etc. while trapped in their homes using the Internet
for everything.<span style="mso-spacerun: yes;"> </span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Big Tech has been underhandedly striving to gather as much
medical data as possible, for years now -- e.g. <a href="https://techcrunch.com/2019/10/22/google-has-used-contract-swaps-to-get-bulk-access-terms-to-nhs-patient-data/">Google
Deep Mind's series of sweetheart deals</a> with the British health system to
garner access to peoples' medical records; or <a href="https://amp.theguardian.com/technology/2019/nov/12/google-medical-data-project-nightingale-secret-transfer-us-health-information">Project
Nightingale</a> which saw Google quietly capture 50 million Americans medical
records.<span style="mso-spacerun: yes;"> </span>Gathering medical data from a
wide population with a view toward pandemic-related analysis and prediction is
absolute golden for Big Tech.<span style="mso-spacerun: yes;"> </span>This data
and the pipelines that bring it their way will continue to yield value for
these companies and their government partners long after COVID-19 has been reduced
to the level of one more irritating seasonal infection.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
As everyone becomes increasingly fearful for the lives of
their elderly friends and relations, centralized monitoring of everybody's
location and movements and physiological data is increasingly taken as a Good
Thing.<span style="mso-spacerun: yes;"> </span>Today uploaded temperature
readings from a million+ wireless digital thermometers are <a href="https://healthcareitnews.com/news/digital-thermometer-data-may-provide-insight-covid-19-surges">letting
us track the spread of COVID-19</a> around the US.<span style="mso-spacerun: yes;"> </span><span style="mso-spacerun: yes;"> </span><a href="https://med.stanford.edu/news/all-news/2017/01/wearable-sensors-can-tell-when-you-are-getting-sick.html">Stanford
researchers have also shown that</a>, by using AI anomaly detection on data
from heart-rate variability, body temperature and pulse oximetry , one can
identify a person is sick even before they show any symptoms.<span style="mso-spacerun: yes;"> </span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
But then what happens when it becomes standard for your
smartwatch, smartphone and fitness tracker to upload your data to Big Tech and
Big Government so they can track and analyze disease spread?<span style="mso-spacerun: yes;"> </span>Do you really trust these corporate and
governmental entities not to use this data for other purposes -- and not to
find ways to quietly keep collecting and utilizing similar data?<span style="mso-spacerun: yes;"> </span><a href="https://www.cnet.com/news/snowden-warns-government-surveillance-amid-covid-19-could-be-long-lasting/">Edward
Snowden has recently gone on record</a> that, no, he does not.<span style="mso-spacerun: yes;"> </span>As you may have guessed,<span style="mso-spacerun: yes;"> </span>I don't either.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Yet <a href="https://news.sky.com/story/coronavirus-govt-set-to-release-contact-tracking-app-which-detects-nearby-virus-carriers-11966243">the
UK is already going directly down this path</a>, with a governmental software
app that detects and tracks nearby COVID-19 sufferers.<span style="mso-spacerun: yes;"> </span>Completely harmless, extremely helpful --
until the same tech and organizational set up is used to track other things of
interest to the ruling politicos and their business and military allies.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Big Brother is watching your heart rate, your temperature
and your blood oxygen level -- better be sure your heart doesn't pound too much
when you walk past that political demonstration, or your credit rating's going
way down!!<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Global monitoring of human movement and human physiology can
do wonders for optimizing global health, during a pandemic and otherwise -- but
it should be done with decentralized, secure tools.<span style="mso-spacerun: yes;"> </span>Otherwise one is placing great trust in the
entities that are gathering and utilizing this data -- not only to do helpful
things with it in the pandemic, but not to leverage this data and related
data-gathering capabilities later in the interest of goals different from that
of global human benefit.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
At the moment most decentralized networks and associated
software tools are still in fairly early states of development -- so to combat
COVID-19 fast we are understandably relying on centralized methods.<span style="mso-spacerun: yes;"> </span>But this will not be the last pandemic nor the
last acute, unprecedented global crisis that humanity faces.<span style="mso-spacerun: yes;"> </span>It is important work so that for the next
such situation that arises, decentralized frameworks will be fully prepared to
play a leading role in helping humanity cope.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Otherwise, each successive crisis will serve to concentrate
more and more wealth and power in the hands of a small elite -- which is not at
all the best way to create a beneficial future for humanity and its technological
children.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<!--EndFragment--><br /></div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com8tag:blogger.com,1999:blog-11168555.post-13175440470609071792020-04-10T04:08:00.000-04:002020-04-10T04:08:17.278-04:00Can We "Discover" Semantic Primitives for Commonsense and Math via Semantic Relation Extraction from Corpora?<div dir="ltr" style="text-align: left;" trbidi="on">
Once more wild-ish train of thought completely unrelated to anything immediately practical … I was thinking about Chalmers’ idea from <a href="https://www.amazon.com/Constructing-World-David-J-Chalmers/dp/019960858X@ca">Constructing the World</a> that the notion of universal semantic primitives underlying all human concepts might be rendered sensible by use of intensional logic … i.e. extensionally reducing all concepts to combinations of a few dozen primitives [plus raw perception/action primitives] is doomed to fail (as shown by eons of pedantic pickery in the analytical philosophy literature) but doing the reduction intensionally seems to basically work….<br />
<br />
But in his book he argues why this is the case and gives lots of examples but doesn’t fully perform the reduction as that’s too big a job (there are a lot of concepts to intensionally reduce…)<br />
<br />
So it occurred to me if we managed to do decent semantic-relation-extraction from large NL corpora, then if Chalmers is right, there would be a set of a few dozen concepts such that doing intensional-logic operations to combine them (plus perception/action primitives) would yield close approximations (small Intensional Difference) from any given concepts<br />
<br />
In vector embedding space, it might mean that any concept can be expressed fairly closely via a combination of the embedding vectors from a few dozen concepts, using combinatory operators like vector sum and pointwise min …<br />
<br />
As I recall it the intensional-combination operators used in Chalmer’s philosophical arguments don’t involve so much advanced quantifier-munging so basic fuzzy-propositional-logic operators might do it…<br />
<br />
Now if we cross-correlate this with Lakoff and Nunez’s thoughts in “<a href="https://www.amazon.com/Where-Mathematics-Come-Embodied-Brings/dp/0465037712">Where Mathematics Comes From?</a>” — where they argue that math theorem proving is done largely by unconscious analogy to reasoning about everyday physical situations — then we get the idea that morphisms from common-sense domains to abstract domains guide math theorem-proving, and that these potential generators of the algebra of commonsense concepts, can be mapped into abstract math-patterns (e.g. math-domain-independent proof strategies/tactics) that serve as generators of proofs for human-friendly mathematics….<br />
<br />
Which led me to wonder if one could form an interesting corpus from videos of math profs going thru proofs online at the whiteboard. One would then capture the verbal explanations along with proofs, hopefully capturing some of the commonsense intuitions/analogies behind the proof steps… from such a corpus one might be able to mine some of the correspondences Lakoff and Nunez wrote about….<br />
<br />
There won’t be a seq2seq model mapping mathematicians’ mutterings into full Mizar proofs, but there could be useful guidance for pruning theorem-prover activity in models of the conceptual flow in mathematician’s proof-accompanying verbalizations.... <br />
<br />
Can we direct proofs from premises to conclusions, via drawing a vector V pointing from the embedding vector for the premise to the embedding vector for the conclusion, and using say the midpoint of V as a subgoal for getting from premise to the conclusion ... and where the basis for the vector space is the primitive mathematical concepts that are the Lakoff-and-Nunez-ian morphic image of primitive everyday-human-world concepts?<br />
<br />
Alas making this sort of thing work is 8 billion times harder than conceptualizing it. But conceptualization is a start ;)<br />
<br />
<br />
<br /></div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com0tag:blogger.com,1999:blog-11168555.post-7578216379168244132020-04-10T03:47:00.000-04:002020-04-10T03:47:03.448-04:00Logical Inference Control via Quantum Partial Search — Maybe<div dir="ltr" style="text-align: left;" trbidi="on">
<br />
Hmmm...<br />
<br />
While running SingularityNET and thinking about next-generation OpenCog and helping Ruiting with our charming little maniac Qorxi are taking up most of my time, I can’t help thinking here and there about quantum AI …<br />
<br />
Quantum computing is moving toward practical realization — it’s still got a long way to go, but clearly the Schrodinger’s cat is out of the bag … the time when every server has a QPU alongside its GPU is now something quite concrete to foresee…<br />
<br />
So I’m thinking a bit about how to use <a href="https://arxiv.org/abs/1709.08230">quantum partial search</a> (Grover's algorithm on a chunked database) to speed up backward-chaining logical inference dramatically. <br />
<br />
Suppose we are searching in some set S for those x in S that satisfying property P. (The interesting case is where S is known implicitly rather than explicitly listed.)<br />
<br />
Suppose we have some distribution f over S, which assigns a probability value f(x) to each element of S — interpretable as the prior probability that x will satisfy P<br />
<br />
Suppose we divide S into bin S1, S2,…, Sk, so that the expected number of x that satisfy P is the same for each Si (in which case the bins containing higher-probability x will have smaller cardinality) …<br />
<br />
Then we can use quantum partial search to find a bin that contains x that satisfies P. <br />
<br />
If the size of S is N and the number of items per bin were constant b, then the time required is (pi/4) sqrt(N/b). Time required increases with uneven-ness of bins (which means non-uniformity of distribution f, in this setup).<br />
<br />
In an inference context, for instance, suppose one has a desired conclusion C and n premises Pi. One wants to know for what combinations Pi * Pj ==> C. One then constructs an N = n^2 dimensional Hilbert space, which has a basis vector corresponding to each combination (i,j). One call to the quantum oracle can tell us whether Pi * Pj ==> C for some particular (i,j) (note though that this call must be implementable as a unitary transformation on the Hilbert space — but following the standard math of quantum circuits it can be set up this way). <br />
<br />
Using straight Grover’s algorithm, one can then find which Pi * Pj ==> C in sqrt(N) time.<br />
<br />
If one wants to leverage the prior distribution, one can find which bin(s) the premise-pairs so that {Pi * Pj ==> C } live in, in time (pi/4) sqrt(c*N/b) where c>1 is the correction for the non-uniformity of the prior and b is the average number of pairs per bin.<br />
<br />
With a uniform prior, one is finding log(N/b) bits of information about what the premises are (and narrowing down to a search over b items).<br />
<br />
With a non-uniform prior, one is still narrowing down *on average* to a search over b items, so is still finding log(N/b) bits on average about where the items are.<br />
<br />
This could be useful e.g. in a hybrid classical-quantum context, where the quantum computer is used to narrow down a very large number of options to a more modest number, which are then searched through using classical methods.<br />
<br />
It could also be useful as a heuristic layer on top of Grover’s algorithm. I.e., one could do this prior-probability-guided search to narrow things down to a bin, and then do full-on Grover’s algorithm within the bin selected.<br />
<br />
Constructing the bins in an artful way, so that e.g. bins tend to have similar entities in them, could potentially make things work even faster. Specifically, if the elements in each bin tend to be similar to each other, then the bin may effectively be a lower-dimensional subspace, which means the algorithm will work faster on that bin. So there would be advantage to clustering the items being searched before constructing the bins. If items that are clustered together tend to have similar prior probabilities, then the bins would tend to be lower-dimensional and things would tend to go faster.<br />
<br />
<b>Grover’s Algorithm and Natural Gradients</b><br />
<br />
Now if we want to go even deeper down the rabbit hole — <a href="https://arxiv.org/pdf/1701.02549.pdf">this funky paper </a>shows that the quantum search problem reduces to finding optimal geodesic paths that minimize lengths on a manifold of pure density matrices with a metric structure defined by the Wigner-Yanase metric tensor …<br />
<br />
Fisher metric geeks will simultaneously drop their jaws in amazement, and nod and grin in a self-satisfied way<br />
<br />
So what we see here is that Grover’s algorithm is actually just following the natural gradient ... well sort of…<br />
<br />
Putting some pieces together … We have seen that partial quantum search (Grover’s algorithm over a chunked database) can be set up to provide rapid (on average) approximate location of an item in an implicit database, where the average is taken relative to a given probability distribution (and the distribution is used to guide the chunking of the database)….<br />
<br />
Well then — this partial quantum search on a database chunked-according-to-a-certain-distribution, should presumably correspond to following the natural gradient on a manifold of pure density matrices with a metric structure conditioned by that same distribution…<br />
<br />
Which — if it actually holds up — is not really all that deep, just connecting some (quantum) dots, but sorta points in a nice quantum AI direction…<br />
<br />
<b>Post-Script: Wow, This Stuff May Be Implementable?</b><br />
<br />
I was amazed/ amused to note some small-scale <a href="https://www.researchgate.net/publication/325436391_The_first_iteration_of_Grover's_algorithm_using_classical_light_with_orbital_angular_momentum">practical implementations of Grover’s Algorithm using Orbital Angular Momentum</a> …<br />
<br />
It’s all classical optics except preparation of the initial state (which is where the Oracle gets packed).<br />
<br />
Could this be how our quantum-accelerated logical inference control is going to work? Quantum optics plugins for the server … or the cortex?<br />
<div>
<br /></div>
</div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com4tag:blogger.com,1999:blog-11168555.post-86104819543209919102019-07-22T00:20:00.002-04:002019-07-22T00:27:30.601-04:00Toward an Abstract Energetics of Computational Processes (in Brains, Minds, Physics and Beyond)<div dir="ltr" style="text-align: left;" trbidi="on">
<div dir="ltr" style="text-align: left;" trbidi="on">
<!--[if !mso]>
<style>
v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
</style>
<![endif]--><!--[if gte mso 9]><xml>
<o:OfficeDocumentSettings>
<o:AllowPNG/>
</o:OfficeDocumentSettings>
</xml><![endif]-->
<!--[if gte mso 9]><xml>
<w:WordDocument>
<w:View>Normal</w:View>
<w:Zoom>0</w:Zoom>
<w:TrackMoves>false</w:TrackMoves>
<w:TrackFormatting/>
<w:PunctuationKerning/>
<w:ValidateAgainstSchemas/>
<w:SaveIfXMLInvalid>false</w:SaveIfXMLInvalid>
<w:IgnoreMixedContent>false</w:IgnoreMixedContent>
<w:AlwaysShowPlaceholderText>false</w:AlwaysShowPlaceholderText>
<w:DoNotPromoteQF/>
<w:LidThemeOther>EN-US</w:LidThemeOther>
<w:LidThemeAsian>JA</w:LidThemeAsian>
<w:LidThemeComplexScript>X-NONE</w:LidThemeComplexScript>
<w:Compatibility>
<w:BreakWrappedTables/>
<w:SnapToGridInCell/>
<w:WrapTextWithPunct/>
<w:UseAsianBreakRules/>
<w:DontGrowAutofit/>
<w:SplitPgBreakAndParaMark/>
<w:EnableOpenTypeKerning/>
<w:DontFlipMirrorIndents/>
<w:OverrideTableStyleHps/>
<w:UseFELayout/>
</w:Compatibility>
<m:mathPr>
<m:mathFont m:val="Cambria Math"/>
<m:brkBin m:val="before"/>
<m:brkBinSub m:val="--"/>
<m:smallFrac m:val="off"/>
<m:dispDef/>
<m:lMargin m:val="0"/>
<m:rMargin m:val="0"/>
<m:defJc m:val="centerGroup"/>
<m:wrapIndent m:val="1440"/>
<m:intLim m:val="subSup"/>
<m:naryLim m:val="undOvr"/>
</m:mathPr></w:WordDocument>
</xml><![endif]--><!--[if gte mso 9]><xml>
<w:LatentStyles DefLockedState="false" DefUnhideWhenUsed="true"
DefSemiHidden="true" DefQFormat="false" DefPriority="99"
LatentStyleCount="276">
<w:LsdException Locked="false" Priority="0" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Normal"/>
<w:LsdException Locked="false" Priority="9" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="heading 1"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 2"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 3"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 4"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 5"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 6"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 7"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 8"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 9"/>
<w:LsdException Locked="false" Priority="39" Name="toc 1"/>
<w:LsdException Locked="false" Priority="39" Name="toc 2"/>
<w:LsdException Locked="false" Priority="39" Name="toc 3"/>
<w:LsdException Locked="false" Priority="39" Name="toc 4"/>
<w:LsdException Locked="false" Priority="39" Name="toc 5"/>
<w:LsdException Locked="false" Priority="39" Name="toc 6"/>
<w:LsdException Locked="false" Priority="39" Name="toc 7"/>
<w:LsdException Locked="false" Priority="39" Name="toc 8"/>
<w:LsdException Locked="false" Priority="39" Name="toc 9"/>
<w:LsdException Locked="false" Priority="35" QFormat="true" Name="caption"/>
<w:LsdException Locked="false" Priority="10" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Title"/>
<w:LsdException Locked="false" Priority="1" Name="Default Paragraph Font"/>
<w:LsdException Locked="false" Priority="11" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Subtitle"/>
<w:LsdException Locked="false" Priority="22" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Strong"/>
<w:LsdException Locked="false" Priority="20" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Emphasis"/>
<w:LsdException Locked="false" Priority="59" SemiHidden="false"
UnhideWhenUsed="false" Name="Table Grid"/>
<w:LsdException Locked="false" UnhideWhenUsed="false" Name="Placeholder Text"/>
<w:LsdException Locked="false" Priority="1" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="No Spacing"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 1"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 1"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 1"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 1"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 1"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 1"/>
<w:LsdException Locked="false" UnhideWhenUsed="false" Name="Revision"/>
<w:LsdException Locked="false" Priority="34" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="List Paragraph"/>
<w:LsdException Locked="false" Priority="29" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Quote"/>
<w:LsdException Locked="false" Priority="30" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Intense Quote"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 1"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 1"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 1"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 1"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 1"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 1"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 1"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 1"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 2"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 2"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 2"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 2"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 2"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 2"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 2"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 2"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 2"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 2"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 2"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 2"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 2"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 2"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 3"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 3"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 3"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 3"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 3"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 3"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 3"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 3"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 3"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 3"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 3"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 3"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 3"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 3"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 4"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 4"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 4"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 4"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 4"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 4"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 4"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 4"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 4"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 4"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 4"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 4"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 4"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 4"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 5"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 5"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 5"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 5"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 5"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 5"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 5"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 5"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 5"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 5"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 5"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 5"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 5"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 5"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 6"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 6"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 6"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 6"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 6"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 6"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 6"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 6"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 6"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 6"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 6"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 6"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 6"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 6"/>
<w:LsdException Locked="false" Priority="19" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Subtle Emphasis"/>
<w:LsdException Locked="false" Priority="21" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Intense Emphasis"/>
<w:LsdException Locked="false" Priority="31" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Subtle Reference"/>
<w:LsdException Locked="false" Priority="32" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Intense Reference"/>
<w:LsdException Locked="false" Priority="33" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Book Title"/>
<w:LsdException Locked="false" Priority="37" Name="Bibliography"/>
<w:LsdException Locked="false" Priority="39" QFormat="true" Name="TOC Heading"/>
</w:LatentStyles>
</xml><![endif]-->
<style>
<!--
/* Font Definitions */
@font-face
{font-family:Arial;
panose-1:2 11 6 4 2 2 2 2 2 4;
mso-font-charset:0;
mso-generic-font-family:auto;
mso-font-pitch:variable;
mso-font-signature:-536859905 -1073711037 9 0 511 0;}
@font-face
{font-family:"MS 明朝";
mso-font-charset:78;
mso-generic-font-family:auto;
mso-font-pitch:variable;
mso-font-signature:-536870145 1791491579 18 0 131231 0;}
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;
mso-font-charset:0;
mso-generic-font-family:auto;
mso-font-pitch:variable;
mso-font-signature:-536870145 1107305727 0 0 415 0;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;
mso-font-charset:0;
mso-generic-font-family:auto;
mso-font-pitch:variable;
mso-font-signature:-520092929 1073786111 9 0 415 0;}
@font-face
{font-family:Cambria;
panose-1:2 4 5 3 5 4 6 3 2 4;
mso-font-charset:0;
mso-generic-font-family:auto;
mso-font-pitch:variable;
mso-font-signature:-536870145 1073743103 0 0 415 0;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{mso-style-unhide:no;
mso-style-qformat:yes;
mso-style-parent:"";
margin:0in;
margin-bottom:.0001pt;
mso-pagination:widow-orphan;
font-size:12.0pt;
font-family:Cambria;
mso-ascii-font-family:Cambria;
mso-ascii-theme-font:minor-latin;
mso-fareast-font-family:"MS 明朝";
mso-fareast-theme-font:minor-fareast;
mso-hansi-font-family:Cambria;
mso-hansi-theme-font:minor-latin;
mso-bidi-font-family:"Times New Roman";
mso-bidi-theme-font:minor-bidi;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;
text-underline:single;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-noshow:yes;
mso-style-priority:99;
color:purple;
mso-themecolor:followedhyperlink;
text-decoration:underline;
text-underline:single;}
.MsoChpDefault
{mso-style-type:export-only;
mso-default-props:yes;
font-family:Cambria;
mso-ascii-font-family:Cambria;
mso-ascii-theme-font:minor-latin;
mso-fareast-font-family:"MS 明朝";
mso-fareast-theme-font:minor-fareast;
mso-hansi-font-family:Cambria;
mso-hansi-theme-font:minor-latin;
mso-bidi-font-family:"Times New Roman";
mso-bidi-theme-font:minor-bidi;}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.25in 1.0in 1.25in;
mso-header-margin:.5in;
mso-footer-margin:.5in;
mso-paper-source:0;}
div.WordSection1
{page:WordSection1;}
</style>
</div>
<a href="https://www.blogger.com/blogger.g?blogID=11168555" imageanchor="1" style="clear: left; float: left; margin-bottom: 1em; margin-right: 1em;"><img border="0" height="640" src="file://localhost/private/var/folders/jj/_pxwy1955jg5yfc2fxz06hzw0000gn/T/TemporaryItems/msoclip/0/clip_image004.png" style="cursor: move;" v:shapes="Picture_x0020_2" width="0" /></a><br />
<div class="MsoNormal">
<a href="https://www.blogger.com/blogger.g?blogID=11168555" imageanchor="1" style="clear: right; float: right; margin-bottom: 1em; margin-left: 1em;"><img border="0" height="640" src="file://localhost/private/var/folders/jj/_pxwy1955jg5yfc2fxz06hzw0000gn/T/TemporaryItems/msoclip/0/clip_image004.png" style="cursor: move;" v:shapes="Picture_x0020_2" width="0" /></a></div>
<div class="MsoNormal">
<span style="font-family: "calibri";">Given the successes of energy-based
formalisms in physics, it is natural to want to extend them into other domains
like computation and cognition.</span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">In this vein: My aim here is to sketch what I
think is a workable approach to an energetics of computational processes
(construed very broadly).<span style="mso-spacerun: yes;"> </span><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">By this I mean: I will explain how one can
articulate highly general principles of the dynamics of computational
processes, that take a similar form to physics principles such as the
stationary action principle (which often takes the form of "least
action") and the Second Law of Thermodynamics (the principle of entropy
non-decrease).<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Why am I interested in this topic?<span style="mso-spacerun: yes;"> </span>Two related reasons, actually.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">First, I would like to create a "General
Theory of General Intelligence" -- or to be more precise, a general theory
of what kinds of systems can display what levels of general intelligence in
what environments given realistically limited (space, time and energy)
resources.<span style="mso-spacerun: yes;"> </span>Marcus Hutter's Universal AI
theory is great but it doesn't say much about general intelligence under
realistic resource assumptions, most of its power is limited to the case of AI
systems with unrealistically<span style="mso-spacerun: yes;"> </span>massive
processing power.<span style="mso-spacerun: yes;"> </span><span style="mso-spacerun: yes;"> </span>I have published some ideas on this before --
e.g. <a href="https://arxiv.org/abs/1703.04361"><span style="mso-bidi-font-family: Helvetica;">formalizing Cognitive Synergy in terms of category theory</span></a>,
and <a href="https://ieeexplore.ieee.org/document/5250687"><span style="mso-bidi-font-family: Helvetica;">articulating the Embodied Communication
Prior</span></a> in regard to which human-like agents attempt to be intelligent
-- but nothing remotely near fully satisfying.<span style="mso-spacerun: yes;">
</span>So I'm searching for new directions.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Second, I would like to come up with a real
scientific theory of psi phenomena.<span style="mso-spacerun: yes;"> </span>I am
inclined toward what I call "euryphysical" theories -- i.e. theories
that involve embedding our 4D spacetime continuum in a larger space (which
could be a higher dimensional space or could be a non-dimensional topological
space of some sort).<span style="mso-spacerun: yes;"> </span>However, this begs
the question of what this large space is like -- what rules govern
"dynamics" in this space?<span style="mso-spacerun: yes;"> </span>In <a href="http://journals.sfu.ca/jnonlocality/index.php/jnonlocality/article/view/65"><span style="mso-bidi-font-family: Helvetica;">my paper on Euryphysics</span></a>, I
give some rough ideas in this direction, but again nothing fully satisfying.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<a href="https://www.blogger.com/blogger.g?blogID=11168555" imageanchor="1" style="clear: right; float: right; margin-bottom: 1em; margin-left: 1em;"><img border="0" height="640" src="file://localhost/private/var/folders/jj/_pxwy1955jg5yfc2fxz06hzw0000gn/T/TemporaryItems/msoclip/0/clip_image004.png" style="cursor: move;" v:shapes="Picture_x0020_2" width="0" /></a><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">It would be nice if mind dynamics -- both in
a traditional AI setting and in a more out-there euryphysical setting -- could
be modeled on dynamical theories in physics, which are based on ideas like
stationary action.<span style="mso-spacerun: yes;"> </span>After all, if as
Peirce said "matter is just mind hide-bound with habit" then perhaps
the laws of matter are in some way simplifications or specializations of the
laws of mind -- and<span style="mso-spacerun: yes;"> </span>maybe there are laws
of mind with roughly analogous form to some of the current laws of physics.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;"><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">A Few Comments on Friston's Free Energy Ideas<o:p></o:p></span></b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Friston's "free energy principle"
represents one well-known effort in the direction of modeling cognition using
physics-ish principles.<span style="mso-spacerun: yes;"> </span>It seems to me
that Friston's ideas have some fundamental shortcomings -- but reviewing these
shortcomings has some value for understanding how to take a more workable
approach.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">I should clarify that my own thinking
described in this blog post was not inspired by Friston's thinking to any
degree, but more so by long-ago reading in the systems-theory literature --
i.e. reading stuff like Ilya Prigogine's <a href="https://www.amazon.com/Order-Out-Chaos-Dialogue-Thinkers/dp/1786631008">Order
out of Chaos</a></span><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">]</span><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;"> and Eric Jantsch's <a href="https://www.amazon.com/Self-Organizing-Universe-Scientific-Implications-Innovations/dp/0080243126">The
Self-Organizing Universe</a> <span style="mso-spacerun: yes;"> </span>and Hermann
Haken's <a href="https://www.amazon.com/Principles-Brain-Functioning-Synergetic-Synergetics/dp/3540589678">Synergetics</a>.<span style="mso-spacerun: yes;"> </span>These authors represented a tradition
within the complex-systems research community, of using far-from-equilibrium
thermodynamics as a guide for thinking about life, the universe and
everything.<span style="mso-spacerun: yes;"> </span><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Friston's "free energy principle"
seems to have a somewhat similar conceptual orientation, but confusingly to me,
doesn't seem to incorporate the lessons of far-from-equilibrium thermodynamics
that thoroughly, being based more on equilibrium-thermodynamics-ish ideas.<span style="mso-spacerun: yes;"> </span><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">I haven't read everything Friston has
written, but have skimmed various papers of his over the years, and recently
looked at the much-discussed papers <a href="https://royalsocietypublishing.org/doi/10.1098/rsif.2017.0792">The Markov
blankets of life: autonomy, active inference and the free energy principle</a><o:p></o:p></span></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">and</span><span style="font-family: "calibri"; font-size: 10.0pt;">
</span><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;"><a href="https://www.frontiersin.org/articles/10.3389/fncom.2018.00090/full">The
Anatomy of Inference: Generative Models and Brain Structure</a></span><span style="font-family: "calibri"; font-size: 10.0pt;"> <o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal" style="text-align: left;">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">My general confusion about Friston's ideas is
largely the same as that expressed by the authors of blog posts such as<o:p></o:p></span></div>
<div class="MsoNormal" style="text-align: left;">
<span style="font-family: "calibri"; font-size: 10.0pt;"><a href="https://www.lesswrong.com/posts/wpZJvgQ4HvJE2bysy/god-help-us-let-s-try-to-understand-friston-on-free-energy">God
Help Us, Let’s Try To Understand Friston On Free Energy</a> and <a href="https://www.umsu.de/wo/2013/600">The lure of free energy</a> </span><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;"><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">As the latter post notes, regarding
perception, Friston basically posits that neural and cognitive systems are
engaged with trying to model the world they live in, and do so by looking for
models with maximum probability conditioned on the data they've observed.<span style="mso-spacerun: yes;"> </span>This is a useful but not adventurous
perceptive, and one can formulate it in terms of trying to find models
with<span style="mso-spacerun: yes;"> </span>minimum KL-divergence to reality,
which is one among many ways to describe Bayesian inference ... and which can
be mathematically viewed as attempting to minimize a certain "free
energy" function.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">Friston then attempts to extend this
principle to action via a notion of "active inference", and this is
where things get dodgier. <span style="mso-spacerun: yes;"> </span>As the
above-linked "Markov Blankets" paper puts it,<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<i style="mso-bidi-font-style: normal;"><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">"Active
inference is a cornerstone of the free energy principle. This principle states
that for organisms to maintain their integrity they must minimize variational
free energy. <span style="mso-spacerun: yes;"> </span>Variational free energy bounds
surprise because the former can be shown to be either greater than or equal to
the latter. It follows that any organism that minimizes free energy thereby
reduces surprise—which is the same as saying that such an organism maximizes
evidence for its own model, i.e. its own existence<o:p></o:p></span></i></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<i style="mso-bidi-font-style: normal;"><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">...<o:p></o:p></span></i></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<i style="mso-bidi-font-style: normal;"><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">This
interpretation means that changing internal states is equivalent to inferring
the most probable, hidden causes of sensory signals in terms of expectations
about states of the environment<o:p></o:p></span></i></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<i style="mso-bidi-font-style: normal;"><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">...<o:p></o:p></span></i></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<i style="mso-bidi-font-style: normal;"><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">[A]
biological system must possess a generative model with temporal depth, which,
in turn, implies that it can sample among different options and select the
option that has the greatest (expected) evidence or least (expected) free
energy. The options sampled from are intuitively probabilistic and future
oriented. Hence, living systems are able to ‘free’ themselves from their
proximal conditions by making inferences about probabilistic future states and
acting so as to minimize the expected surprise (i.e. uncertainty) associated
with those possible future states. This capacity connects biological qua
homeostatic systems with autonomy, as the latter denotes an organism’s capacity
to regulate its internal milieu in the face of an ever-changing environment.
This means that if a system is autonomous it must also be adaptive, where
adaptivity refers to an ability to operate differentially in certain
circumstances. <o:p></o:p></span></i></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<i style="mso-bidi-font-style: normal;"><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">...<o:p></o:p></span></i></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<i style="mso-bidi-font-style: normal;"><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">The
key difference between mere and adaptive active inference rests upon selecting
among different actions based upon deep (temporal) generative models that
minimize the free energy expected under different courses of action.<o:p></o:p></span></i></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<i style="mso-bidi-font-style: normal;"><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">This
suggests that living systems can transcend their immediate present state and
work towards occupying states with a free energy minimum."<o:p></o:p></span></i></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">If you are a math/physics oriented person
and find the above quotes frustratingly vague, unfortunately you will find that
the rest of the paper is equally vague on the confusing points, and Friston's
other papers are also.<span style="mso-spacerun: yes;"> </span><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">What it sounds like to me (doing some
"active inference" myself to try to understand what the paper is
trying to say) is that active inference is being portrayed as a process by
which cognitive systems take actions aimed at putting themselves in situations
that will be minimally surprising, i.e. in which they will have the most
accurate models of reality.<span style="mso-spacerun: yes;"> </span>If taken
literally this cannot be true, as it would predict that intelligent systems
systematically seek simpler situations they can model better -- which is
obviously not a full description of human motivation, for instance.<span style="mso-spacerun: yes;"> </span>We do have a motivation to put ourselves in
comprehensible, accurately model-able situations -- but we also have other
motivations, such as the desire to perceive novelty and to challenge ourselves,
which sometimes contradict our will to have a comprehensible environment.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">The main thing that jumps out at me
when reading what Friston and colleagues write about active inference is that
it's too much about states and not enough about paths.<span style="mso-spacerun: yes;"> </span>To model far-from-equilibrium thermodynamics
using energy-based formalisms, one needs to think about paths and path
entropies and such, not just about things like "<i style="mso-bidi-font-style: normal;"> work[ing] towards occupying states with a free energy minimum."</i><span style="mso-spacerun: yes;"> </span>Instead of thinking about ideas like "<i style="mso-bidi-font-style: normal;"> </i>selecting among different actions based
upon deep (temporal) generative models that minimize the free energy expected
under different courses of action." in terms of states with free
energy<span style="mso-spacerun: yes;"> </span>minimum, one needs to be thinking
about action selection in terms of stationarity of action functions evaluated
along multiple paths.<o:p></o:p></span></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; font-size: 10.0pt;"><span style="mso-spacerun: yes;"> </span><o:p></o:p></span></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;"><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Energetics for Far-From-Equilibrium Thermodynamics<o:p></o:p></span></b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">It seems clear that equilibrium
thermodynamics isn’t really what we want to use as a guide for cognitive
information processing.<span style="mso-spacerun: yes;"> </span>Fortunately, the
recent thermodynamics literature contains some quite interesting results
regarding path entropy in far-from-equilibrium thermodynamics.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Abaimov's paper <span style="mso-spacerun: yes;"> </span><a href="https://arxiv.org/pdf/0906.0190.pdf">General
formalism of non-equilibrium statistical mechanics, path approach</a> </span><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">and
Raphael Chetrite and Hugo Touchette's paper <a href="https://hal.archives-ouvertes.fr/hal-00915479/document">Nonequilibrium
Microcanonical and Canonical Ensembles and Their Equivalence</a></span><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;"> each tell part of the story.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">David Rogers and Susan Rempe in </span><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;"><a href="https://arxiv.org/pdf/1105.5619.pdf">A First and Second Law for
Nonequilibrium Thermodynamics: Maximum Entropy Derivation of the
Fluctuation-Dissipation Theorem and Entropy Production Functionals</a>"
describe</span><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;"> explicitly the far from equilibrium “path
free energy”, but only for the case of processes with short memory, i.e. state
at time i+1 depends on state i but not earlier ones (which is often fine but
not totally general).<span style="mso-spacerun: yes;"> </span><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">The following table from Rogers and Rempe
summarizes some key points concisely.</span><span style="font-family: "calibri"; font-size: 10.0pt;"><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br />
<br />
<div class="separator" style="clear: both; text-align: center;">
<a href="https://1.bp.blogspot.com/-xMfgB-FIvhM/XTU4YfQFwjI/AAAAAAAAWJU/wC1ypApcyGYTdNG1jJBkljYNoLHIEr5LwCLcBGAs/s1600/Table1.png" imageanchor="1" style="clear: left; float: left; margin-bottom: 1em; margin-left: 1em;"><img border="0" data-original-height="363" data-original-width="900" height="257" src="https://1.bp.blogspot.com/-xMfgB-FIvhM/XTU4YfQFwjI/AAAAAAAAWJU/wC1ypApcyGYTdNG1jJBkljYNoLHIEr5LwCLcBGAs/s640/Table1.png" width="640" /></a></div>
</div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Conceptually, the key point is that we need
to think not about the entropy of a state, but about the "caliber" of
a path -- a normalization of the number of ways that path can be realized.<span style="mso-spacerun: yes;"> </span>This then leads to the notion of the free
energy of a certain path.<span style="mso-spacerun: yes;"> </span></span><span style="font-family: "calibri"; font-size: 10.0pt;"><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">It follows from this body of work that ideas
like "free energy minimization" need to be re-thought dynamically
rather than statically.<span style="mso-spacerun: yes;"> </span>One needs to
think about systems as following paths with differential probability based on
the corresponding path free energies.<span style="mso-spacerun: yes;">
</span>This is in line with the "<a href="https://arxiv.org/abs/1711.03450">Maximum
Caliber principle</a>"<span style="mso-spacerun: yes;"> </span>which is a
generalization of the Maximum Entropy principle to dynamical systems (both
first proposed in clear form by E.T. Jaynes, though Maximum Entropy has been
more widely developed than Maximum Caliber so far).<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">Extending these notions further, Diego
Gonzalez outlines <a href="https://arxiv.org/abs/1404.3249">a Hamiltonian
formalism that is equivalent to path entropy maximization</a>, building on math
from his earlier paper<span style="mso-spacerun: yes;"> </span><a href="https://www.researchgate.net/publication/275336459_Inference_of_trajectories_over_a_time-dependent_phase_space_distribution">Inference
of trajectories over a time-dependent phase space distribution</a>.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;"><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Action Selection and Active Inference<o:p></o:p></span></b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Harking back to Friston for a moment, it
follows that the dynamics of an intelligent system, should be viewed, not as an
attempt by an intelligent system to find a state with minimum free energy or
surprisingness etc., but rather as a process of a system evolving dynamically
along paths chosen probabilistically to have stationary path free energy.<span style="mso-spacerun: yes;"> </span><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">But <span style="mso-spacerun: yes;"> </span>of
course, this would be just as true for an unintelligent system as for an
intelligent system -- it's not a principle of intelligence but just a
restatement of how physics works (in far from equilibrium cases; in equilibrium
cases one can collapse paths to states).<span style="mso-spacerun: yes;">
</span><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">If we want to say something unique about
intelligent systems in this context, we can look at the goals that an
intelligent system is trying to achieve.<span style="mso-spacerun: yes;">
</span>We may say that, along each potential path of the system's evolution,
its various goals will be achieved to a certain degree.<span style="mso-spacerun: yes;"> </span>The system then has can be viewed to have a
certain utility distribution across paths -- some paths are more desirable to
it than others.<span style="mso-spacerun: yes;"> </span>A guiding principle of
action selection would then be: <b style="mso-bidi-font-weight: normal;"><i style="mso-bidi-font-style: normal;">To take an action A so that, conditioned on
action A, the predicted probability distribution across paths is as close as
possible to the distribution implied by the system's goals.</i></b><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">This principle of action selection can be
formalized as KL-divergence minimization if one wishes, and in that sense it
can be formulated as a "free energy minimization" principle.<span style="mso-spacerun: yes;"> </span>But it's a "free energy" defined
across ensembles of paths, not across states.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">A side note is, it's important to understand
that the desirability of a path to an intelligent system need not be
expressible as the expected future utility at all moments of time along that
path.<span style="mso-spacerun: yes;"> </span>The desirability of a path may be
some more holistic function of everything that happens along that path.<span style="mso-spacerun: yes;"> </span>Considering only expected utility as a form
of goal leads to various pathologies related to wireheading, as I argued in <a href="http://multiverseaccordingtoben.blogspot.com/2009/05/reinforcement-learning-some-limitations.html">a
long-ago blog post on ultimate orgasms and such</a>.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;"><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Algorithmic Thermodynamics<o:p></o:p></span></b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Now let's dig a little deeper.<span style="mso-spacerun: yes;"> </span>Can we apply these same ideas beyond the
realm of physics, to more general types of processes that change over time?<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">I am inspired by a general Whiteheadean
notion of procesess as fundamental things.<span style="mso-spacerun: yes;">
</span>However, to keep things concrete, for now I'm going to provisionally
assume that the "processes" involved can be formulated as computer
programs, in some standard Turing-equivalent framework, or maybe a
quantum-computing framework.<span style="mso-spacerun: yes;"> </span>I think
the same ideas actually apply more broadly, but -- one step at a time...<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Let us start with Kohtaro Tadaki's truly
beautiful, simple, elegant paper titled</span><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;"> <a href="https://arxiv.org/pdf/0801.4194.pdf">A statistical mechanical
interpretation of algorithmic information theory</a> </span><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;"><span style="mso-spacerun: yes;"> </span><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Section 6 of Tadaki outlines a majorly aesthetic,
obvious-in-hindsight parallel between algorithmic information theory and
equilibrium thermodynamics.<span style="mso-spacerun: yes;"> </span>There is
seen to be a natural mapping between temperature in thermodynamics and
compression ratio in algorithmic information theory.<span style="mso-spacerun: yes;"> </span>A natural notion of "algorithmic free
energy"<span style="mso-spacerun: yes;"> </span>is formulated, as a sort of
weighted program-length over all possible computer programs (where the weights
depend on the temperature).<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">The following table (drawn from <a href="http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.520.3773&rep=rep1&type=pdf">Tadaki's
presentation here</a>) summarizes the key<span style="mso-spacerun: yes;">
</span>mappings in Tadaki's theory</span><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;"><o:p></o:p></span></div>
<div class="MsoNormal">
<br />
<br />
<a href="https://1.bp.blogspot.com/-8jgW6JKp1j8/XTU4Yf4f6tI/AAAAAAAAWJY/RlNFa0voBHgg3FvV17qoBusTrsJudGo7ACLcBGAs/s1600/Table2.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em; text-align: center;"><img border="0" data-original-height="409" data-original-width="900" height="290" src="https://1.bp.blogspot.com/-8jgW6JKp1j8/XTU4Yf4f6tI/AAAAAAAAWJY/RlNFa0voBHgg3FvV17qoBusTrsJudGo7ACLcBGAs/s640/Table2.png" width="640" /></a><br />
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin; mso-no-proof: yes;"><!--[if gte vml 1]><v:shape
id="Picture_x0020_2" o:spid="_x0000_i1025" type="#_x0000_t75" style='width:6in;
height:196pt;visibility:visible;mso-wrap-style:square'>
<v:imagedata src="file://localhost/private/var/folders/jj/_pxwy1955jg5yfc2fxz06hzw0000gn/T/TemporaryItems/msoclip/0/clip_image003.png"
o:title=""/>
</v:shape><![endif]--><!--[if !vml]--><!--[endif]--></span><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;"><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-bidi-font-family: "Times New Roman"; mso-fareast-font-family: "Times New Roman"; mso-hansi-theme-font: major-latin;">To ground the mappings he outlines,
Tadaki gives a<span style="mso-spacerun: yes;"> </span>simple statistical
mechanical interpretation to algorithmic information theory.<span style="mso-spacerun: yes;"> </span>He models an optimal computer as decoding
equipment at the receiving end of a noiseless binary communication
channel.<span style="mso-spacerun: yes;"> </span>In this context, he regards
programs for this computer as codewords (finite binary strings) and regards
computation results (also finite binary strings) as decoded “symbols.”<span style="mso-spacerun: yes;"> </span>For simplicity he assumes that the infinite
binary string sent through the channel -- constituting a series of codewords in
a prefix-free code is generated by infinitely repeated tosses of a fair
coin.<span style="mso-spacerun: yes;"> </span>Based on this simple reductive
model, Tadaki formulates computation-theoretic analogues to core constructs of
traditional equilibrium thermodynamics.<span style="mso-spacerun: yes;">
</span><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Now let's start putting some pieces together.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Perhaps the most useful observation I will
make in this blog post is:<span style="mso-spacerun: yes;"> </span><b style="mso-bidi-font-weight: normal;"><span style="mso-spacerun: yes;"> </span>It
seems one could port the path-entropy based treatment of far-from-equilibrium
thermodynamics (as seen in the papers I've linked above) to Tadaki's
algorithmic-information context, by looking at sources emitting bits that are
not independent of each other but rather have some probabilistic dependencies..</b><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">By doing so, one would obtain an “algorithmic
energy” function that measures the energy of an algorithmic process over a
period of time -- without assuming that it’s a memoryless process like Tadaki
does in his paper.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">To get this to work, so far as I can limn
without doing all the math (which I don't have time for at the moment, alas), one
needs to assume that the knowledge one has of the dependencies among the bits
produced by the process is given the form of expectations…<span style="mso-spacerun: yes;"> </span>e.g. that we know the average value of
f_k(x_{i+1}, x_i} for various observables f_k ….<span style="mso-spacerun: yes;"> </span>Plus one needs to make some other slightly
funny assumptions that are probably replaceable (the paper assumes “the number
of possible transitions does not depend on the starting point”… but I wonder if
this could be replaced by some assumption about causality…)<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">If I'm not mistaken, this should give us
something like Friston’s free energy principle that actually works and has
meaning….<span style="mso-spacerun: yes;"> </span>I.e. we have a rigorous sense
in which complex algorithmic systems are minimizing free energy.<span style="mso-spacerun: yes;"> </span>The catch is that it’s an algorithmic path
energy -- but hey...<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">More precisely, relative to an observer S who
is observing a system S1 in a<span style="mso-spacerun: yes;"> </span>certain
way (by tabulating conditional probabilities of “how often some event of type A
occurs at time T+s, given some event of type B occurred at time T”) … we may
say the evolution of S1 in S’s perspective obeys an energy minimization
principle, where energy is defined algorithmic-informationally (following my
proposed, not-yet-fleshed-out non-equilibrium generalization of Tadaki’s
approach)…<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;"><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Into the Quantum Rabbit Hole...<o:p></o:p></span></b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Now that we've gone this far, we may as well
plunge in a bit deeper, right?<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Tadaki deals w/ classical computers but --
gesticulating only moderately wildly -- it seems one could generalize his
approach to quantum computers OK.<span style="mso-spacerun: yes;"> </span><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Then one is looking at series of qubits
rather than bits, and instead of tabulating conditional probabilities one is
tabulating amplitudes.<span style="mso-spacerun: yes;"> </span><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">The maximum entropy principle is replaced with
the <a href="https://arxiv.org/abs/1311.0813">stationary quantropy principle</a>
and one still has the situation that: Relative to S who is observing S1 using
some standard linear quantum observables, S1 may be said to evolve according to
a stationary quantropy trajectory, where quantropy is here defined via
generalizing the non-equilibrium generalization of Tadaki’s
algorithmic-informational entropy via replacing the real values w/ complex
values</span><span style="font-family: "calibri"; font-size: 10.0pt;"><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">So we may well get a kind of free-energy
principle for quantum systems also.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">If we want to model cognitive stuff using
bits or qubits, then we have here a physics-ish theory of cognitive
stuff….<span style="mso-spacerun: yes;"> </span>Or at least a sketch of the
start of one…<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;"><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Out Toward the Eurycosm<o:p></o:p></span></b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">One of the motivations for these
investigations was some discussions on higher-dimensional and more broadly
eurycosmic models of psi.<span style="mso-spacerun: yes;"> </span>If there are
non-physical dimensions that connect spatiotemporally distant entities, then
what are the dynamical laws in these dimensions?<span style="mso-spacerun: yes;"> </span>If we can model them as information
dimensions, then maybe the dynamics should be modeled as I’m alluding here…<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Physics dynamics should be recoverable as a
special case of algorithmic-information dynamics where one adds special constraints.<span style="mso-spacerun: yes;"> </span>I.e. the constraints posed by spatial
structure and special relatively etc. should reflect themselves in the
conditional probabilities observed btw various classes of bits or qubits.<span style="mso-spacerun: yes;"> </span><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Then the linear symmetries of spacetime
structure should mean that when you calculate
maximum-path-algorithmic-information distributions relative to these physics
constraints, you end up getting maximum-Shannon-path-entropy
distributions.<span style="mso-spacerun: yes;"> </span>Because macrophysics
results from doing computing using ensembles of randomly chosen computer
programs (i.e. chosen subject to given constraints…).<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">Suppose we want to model a eurycosm that
works according to a principle like Peirce's "tendency to take
habits" <a href="http://multiverseaccordingtoben.blogspot.com/2013/06/physicist-lee-smolin-rediscovers.html">aka
Smolin's Precedence Principle aka Sheldrake's morphic resonance</a>?<span style="mso-spacerun: yes;"> </span>Well then, one can assume that the
probability distribution underlying the emanation of codewords in Tadaki's
model obeys this sort of principle.<span style="mso-spacerun: yes;">
</span>I.e., one can assume that the prior probability of a certain subsequence
is higher if that subsequence, or another subsequence with some of the same
patterns in that sequence, have occurred earlier in the overall sequence.<span style="mso-spacerun: yes;"> </span>Of course there are many ways to modify
Tadaki's precise computational model, and many ways to formalize the notion
that "subsequences with historically more frequent patterns should be more
frequent going forward."<span style="mso-spacerun: yes;"> </span>But
conceptually this is quite straightforward.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">One is forced however to answer the following
question.<span style="mso-spacerun: yes;"> </span>Suppose we assume that the
probability of pattern P occurring in a subsequence beginning at time T is in
some way proportional to the intensity with which P has occurred as a pattern
in the subsequence prior to time T.<span style="mso-spacerun: yes;">
</span>What language of processes are we using to formalize the patterns
P?<span style="mso-spacerun: yes;"> </span>If -- in line with the framework I
articulate in <a href="https://www.amazon.com/Hidden-Pattern-Patternist-Philosophy-Mind/dp/1581129890">The
Hidden Pattern</a> and elsewhere -- we formalize a pattern P in X as a process
P that produces X and is simpler than X -- what is the language in which patterns
are expressed?<span style="mso-spacerun: yes;"> </span><span style="mso-spacerun: yes;"> </span>What is the implicit programming language of
our corner of the eurycosm?<span style="mso-spacerun: yes;"> </span><o:p></o:p></span><br />
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;"><span style="mso-spacerun: yes;"><br /></span></span>
<a href="https://www.blogger.com/blogger.g?blogID=11168555" imageanchor="1" style="clear: right; float: right; margin-bottom: 1em; margin-left: 1em;"><img border="0" height="640" src="file://localhost/private/var/folders/jj/_pxwy1955jg5yfc2fxz06hzw0000gn/T/TemporaryItems/msoclip/0/clip_image004.png" style="cursor: move;" v:shapes="Picture_x0020_2" width="0" /></a><span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;"><span style="mso-spacerun: yes;">For simplicity I have been following Tadaki's conventional Turing machine based computational models here -- with a brief gesture toward quantum computing -- but of course the broad approach outlined here goes beyond these computing paradigms. What if we ported Tadaki's ideas to series of bits emanated by, say, a hypercomputer like the <a href="https://www.sciencedirect.com/science/article/pii/S0304397505009011">Zeno Machine</a>? Then we don't just get a single infinite bit string as output, but a more complex ordinal construction with infinite bit strings of infinite bit strings etc. -- but the math could be worked. If the size of a Zeno Machine program can be quantified by a single real number, then one can assess Zeno Machine programs as patterns in data, and one can define concepts like compression ratio and algorithmic entropy and energy. The paradigm sketched here is not tied to a Turing Machine model of eurycosmic processes, though TMs are certainly easier for initial sketches and calculations than ZMs or even weirder things.</span></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;">I have definitely raised more questions than
I've answered in this long and winding blog post.<span style="mso-spacerun: yes;"> </span>My goal has been to indicate a direction for
research and thinking, one that seems not a huge leap from the current state of
research in various fields, but perhaps dramatic in its utility and implications.<o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<span style="font-family: "calibri"; mso-ascii-theme-font: major-latin; mso-hansi-theme-font: major-latin;"><span style="mso-spacerun: yes;"> </span><o:p></o:p></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<!--EndFragment--><br /></div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com21tag:blogger.com,1999:blog-11168555.post-84778674065350785122019-07-14T12:50:00.000-04:002019-07-14T23:57:41.572-04:00"Deriving" Quantum Logic from Reason, Beauty and Freedom <div dir="ltr" style="text-align: left;" trbidi="on">
<b>One Basic Principle of this Corner of the <a href="http://eurycosm.blogspot.com/">Eurycosm</a>: The Universe Maximizes Freedom Given Constraints of Reason and Beauty</b><br />
<br />
I was musing a bit about the basic concept at the heart of quantum logic and quantum probability: That a particular observer, when reasoning about properties of a system that it cannot in principle ever observe, should use quantum logic / quantum probabilities instead of classical ones.<br />
<br />
I kept wondering: Why should this be the case?<br />
<br />
Then it hit me: It’s just the maximum-entropy principle on a higher level!<br />
<br />
The universe tends to maximize entropy/uncertainty as best it can given the imposed constraints. And quantum amplitude (complex probability) distributions are in a way “more uncertain” than classical ones. So if the universe is maximizing entropy it should be using quantum probabilities wherever possible.<br />
<br />
A little more formally, let’s assume that an observer should reason about their (observable or indirectly assessable) universe in a way that is:<br />
<br />
<br />
<ol style="text-align: left;">
<li>logically consistent: the observations made at one place or time should be logically consistent with those made at other places and times</li>
<li>pleasantly symmetric: the ways uncertainty and information values are measured should obey natural-seeming symmetries, as laid out e.g. by Knuth and Skilling in their paper on Foundations of Inference [<a href="https://arxiv.org/abs/1008.4831">https://arxiv.org/abs/1008.4831</a>] , and followup work on quantum inference [<a href="https://arxiv.org/abs/1712.09725">https://arxiv.org/abs/1712.09725</a>]</li>
<li>maximally entropic: having maximum uncertainty given other imposed constraints. Anything else is assuming more than necessary. This is basically an Occam’s Razor type assumption.</li>
</ol>
<br />
<br />
(I note that logical consistency is closely tied with the potential for useful abstraction. In an inconsistent perceived/modeled world, one can't generalize via the methodology of making a formal abstraction and then deriving implications of that formal abstraction for specific situations (because one can't trust the "deriving" part).... In procedural terms, if a process (specified in a certain language L) starting from a certain seed produces a certain result, then we need it still to be the case later and elsewhere that the same process from the same seed will generate the same result ... if that doesn't work then "pattern recognition" doesn't work so well.... So this sort of induction involving patterns expressed in a language L appears equivalent to logical consistency according to Curry-Howard type correspondences.)<br />
<br />
To put the core point more philosophico-poetically, these three assumptions basically amount to declaring that an observer’s subjective universe should display the three properties of:<br />
<br />
<br />
<ol style="text-align: left;">
<li>Reason</li>
<li>Beauty</li>
<li>Freedom</li>
</ol>
<br />
<br />
Who could argue with that?<br />
<br />
How do reason, beauty and freedom lead to quantum logic?<br />
<br />
I’m short on time as usual so I’m going to run through this pretty fast and loose. Obviously all this needs to be written out much more rigorously, and some hidden rocks may emerge. Let’s just pretend we’re discussing this over a beer and a joint with some jazz in the background…<br />
<br />
We know basic quantum mechanics can be derived from a principle of stationary quantropy (complex valued entropy) [<a href="https://arxiv.org/abs/1311.0813">https://arxiv.org/abs/1311.0813</a>], just as basic classical physics can be derived from a principle of stationary entropy …<br />
<br />
Quantropy ties in naturally with Youssef’s complex-valued truth values [<a href="https://arxiv.org/abs/hep-th/0110253">https://arxiv.org/abs/hep-th/0110253</a>], though one can also interpret/analyze it otherwise…<br />
<br />
It seems clear that modeling a system using complex truth values in a sense reflects MORE uncertainty than modeling a system using real truth values. What I mean is: The complex truth values allow system properties to have the same values they would if modeled w/ real truth values, but also additional values.<br />
<br />
Think about the double-slit experiment: the quantum case allows the electrons to hit the same spots they would in the classical case, but also other spots.<br />
<br />
On the whole, there will be greater logical entropy [<a href="https://arxiv.org/abs/1902.00741">https://arxiv.org/abs/1902.00741</a>] for the quantum case than the classical case, i.e. the percentage of pairs of {property-value assignments for the system} that are considered different will be greater. Double-slit experiment is a clear example here as well.<br />
<br />
So, suppose we had the meta-principle: When modeling any system’s properties, use an adequately symmetric information-theoretic formalism that A) maximizes uncertainty in one’s model of the system, B) will not, in any possible future reality, lead to logical contradictions with future observations.<br />
<br />
By these principles — Reason, Beauty and Freedom — one finds that<br />
<br />
<br />
<ul style="text-align: left;">
<li>for systems properties whose values cannot in principle be observed by you, you should use quantum logic, complex truth values, etc. in preference to regular probabilities (because these have greater uncertainty and there is no problematic contradiction here)</li>
<li>for system properties whose values CAN in principle be observed by you, you can’t use the complex truth values because in the possible realities where you observe the system state, you may come up with conclusions that would contradict some of the complex truth-value assignments</li>
</ul>
<br />
<br />
(E.g. in the double-slit experiment, in the cases where you can in principle observe the electron paths, the quantum assumptions can’t be used as they will lead to conclusions contradictory to observation…)<br />
<br />
A pending question here is why not use quaternionic or octonionic truth values, which Youssef shows also display many of the pleasant symmetries needed to provide reasonable measure of probability and information. The answer has to be that these lack some basic symmetry properties we need to have a workable universe…. This seems plausibly true but needs more detailed elaboration…<br />
<br />
So from the three meta-principles<br />
<br />
<br />
<ol style="text-align: left;">
<li>logical consistency of our models of the world at various times</li>
<li> measurement of uncertainty according to a formalism obeying certain nice symmetry axioms</li>
<li> maximization of uncertainty in our models, subject to the constraints of our observation</li>
</ol>
<br />
<br />
we can derive the conclusion that quantum logic / complex probability should be used for those things an observer in principle can’t measure, whereas classical real probability should be used for those things they can…<br />
<br />
That is, some key aspects our world seem to be derivable from the principle that:<b> The Universe Maximizes Freedom Given Constraints of Reason and Beauty</b><br />
<br />
What is the use of this train of thought?<br />
<br />
I’m not sure yet. But it seems interesting to ground the peculiarity of quantum mechanics in something more fundamental.<br />
<br />
The weird uncertainty of quantum mechanics may seem a bit less weird if one sees it as coming from a principle of <i>assuming the maximum uncertainty one can</i>, consistent with principles of consistency and symmetry. <br />
<br />
Assuming the maximum uncertainty one can, is simply a matter of not assuming more than is necessary. Which seems extremely natural — even if some of its consequences, like quantum logic, can seem less than natural if (as evolution has primed us humans to do) you bring the wrong initial biases to thinking about them.<br />
<br /></div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com6tag:blogger.com,1999:blog-11168555.post-12458589597307858412019-07-09T03:55:00.001-04:002019-07-14T02:52:09.564-04:00The Simulation Hypothesis -- Not Nearly Crazy Enough to Be True<div dir="ltr" style="text-align: left;" trbidi="on">
<div dir="ltr" style="text-align: left;" trbidi="on">
<!--[if gte mso 9]><xml>
<o:OfficeDocumentSettings>
<o:AllowPNG/>
</o:OfficeDocumentSettings>
</xml><![endif]-->
<!--[if gte mso 9]><xml>
<w:WordDocument>
<w:View>Normal</w:View>
<w:Zoom>0</w:Zoom>
<w:TrackMoves/>
<w:TrackFormatting/>
<w:PunctuationKerning/>
<w:ValidateAgainstSchemas/>
<w:SaveIfXMLInvalid>false</w:SaveIfXMLInvalid>
<w:IgnoreMixedContent>false</w:IgnoreMixedContent>
<w:AlwaysShowPlaceholderText>false</w:AlwaysShowPlaceholderText>
<w:DoNotPromoteQF/>
<w:LidThemeOther>EN-US</w:LidThemeOther>
<w:LidThemeAsian>JA</w:LidThemeAsian>
<w:LidThemeComplexScript>X-NONE</w:LidThemeComplexScript>
<w:Compatibility>
<w:BreakWrappedTables/>
<w:SnapToGridInCell/>
<w:WrapTextWithPunct/>
<w:UseAsianBreakRules/>
<w:DontGrowAutofit/>
<w:SplitPgBreakAndParaMark/>
<w:EnableOpenTypeKerning/>
<w:DontFlipMirrorIndents/>
<w:OverrideTableStyleHps/>
<w:UseFELayout/>
</w:Compatibility>
<m:mathPr>
<m:mathFont m:val="Cambria Math"/>
<m:brkBin m:val="before"/>
<m:brkBinSub m:val="--"/>
<m:smallFrac m:val="off"/>
<m:dispDef/>
<m:lMargin m:val="0"/>
<m:rMargin m:val="0"/>
<m:defJc m:val="centerGroup"/>
<m:wrapIndent m:val="1440"/>
<m:intLim m:val="subSup"/>
<m:naryLim m:val="undOvr"/>
</m:mathPr></w:WordDocument>
</xml><![endif]--><!--[if gte mso 9]><xml>
<w:LatentStyles DefLockedState="false" DefUnhideWhenUsed="true"
DefSemiHidden="true" DefQFormat="false" DefPriority="99"
LatentStyleCount="276">
<w:LsdException Locked="false" Priority="0" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Normal"/>
<w:LsdException Locked="false" Priority="9" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="heading 1"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 2"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 3"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 4"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 5"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 6"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 7"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 8"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 9"/>
<w:LsdException Locked="false" Priority="39" Name="toc 1"/>
<w:LsdException Locked="false" Priority="39" Name="toc 2"/>
<w:LsdException Locked="false" Priority="39" Name="toc 3"/>
<w:LsdException Locked="false" Priority="39" Name="toc 4"/>
<w:LsdException Locked="false" Priority="39" Name="toc 5"/>
<w:LsdException Locked="false" Priority="39" Name="toc 6"/>
<w:LsdException Locked="false" Priority="39" Name="toc 7"/>
<w:LsdException Locked="false" Priority="39" Name="toc 8"/>
<w:LsdException Locked="false" Priority="39" Name="toc 9"/>
<w:LsdException Locked="false" Priority="35" QFormat="true" Name="caption"/>
<w:LsdException Locked="false" Priority="10" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Title"/>
<w:LsdException Locked="false" Priority="1" Name="Default Paragraph Font"/>
<w:LsdException Locked="false" Priority="11" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Subtitle"/>
<w:LsdException Locked="false" Priority="22" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Strong"/>
<w:LsdException Locked="false" Priority="20" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Emphasis"/>
<w:LsdException Locked="false" Priority="59" SemiHidden="false"
UnhideWhenUsed="false" Name="Table Grid"/>
<w:LsdException Locked="false" UnhideWhenUsed="false" Name="Placeholder Text"/>
<w:LsdException Locked="false" Priority="1" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="No Spacing"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 1"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 1"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 1"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 1"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 1"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 1"/>
<w:LsdException Locked="false" UnhideWhenUsed="false" Name="Revision"/>
<w:LsdException Locked="false" Priority="34" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="List Paragraph"/>
<w:LsdException Locked="false" Priority="29" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Quote"/>
<w:LsdException Locked="false" Priority="30" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Intense Quote"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 1"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 1"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 1"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 1"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 1"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 1"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 1"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 1"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 2"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 2"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 2"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 2"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 2"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 2"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 2"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 2"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 2"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 2"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 2"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 2"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 2"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 2"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 3"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 3"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 3"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 3"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 3"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 3"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 3"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 3"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 3"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 3"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 3"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 3"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 3"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 3"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 4"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 4"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 4"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 4"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 4"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 4"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 4"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 4"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 4"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 4"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 4"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 4"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 4"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 4"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 5"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 5"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 5"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 5"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 5"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 5"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 5"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 5"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 5"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 5"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 5"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 5"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 5"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 5"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 6"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 6"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 6"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 6"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 6"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 6"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 6"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 6"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 6"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 6"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 6"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 6"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 6"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 6"/>
<w:LsdException Locked="false" Priority="19" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Subtle Emphasis"/>
<w:LsdException Locked="false" Priority="21" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Intense Emphasis"/>
<w:LsdException Locked="false" Priority="31" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Subtle Reference"/>
<w:LsdException Locked="false" Priority="32" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Intense Reference"/>
<w:LsdException Locked="false" Priority="33" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Book Title"/>
<w:LsdException Locked="false" Priority="37" Name="Bibliography"/>
<w:LsdException Locked="false" Priority="39" QFormat="true" Name="TOC Heading"/>
</w:LatentStyles>
</xml><![endif]-->
<style>
<!--
/* Font Definitions */
@font-face
{font-family:"MS 明朝";
panose-1:0 0 0 0 0 0 0 0 0 0;
mso-font-charset:128;
mso-generic-font-family:roman;
mso-font-format:other;
mso-font-pitch:fixed;
mso-font-signature:1 134676480 16 0 131072 0;}
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;
mso-font-charset:0;
mso-generic-font-family:auto;
mso-font-pitch:variable;
mso-font-signature:-536870145 1107305727 0 0 415 0;}
@font-face
{font-family:Cambria;
panose-1:2 4 5 3 5 4 6 3 2 4;
mso-font-charset:0;
mso-generic-font-family:auto;
mso-font-pitch:variable;
mso-font-signature:-536870145 1073743103 0 0 415 0;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{mso-style-unhide:no;
mso-style-qformat:yes;
mso-style-parent:"";
margin:0in;
margin-bottom:.0001pt;
mso-pagination:widow-orphan;
font-size:12.0pt;
font-family:Cambria;
mso-ascii-font-family:Cambria;
mso-ascii-theme-font:minor-latin;
mso-fareast-font-family:"MS 明朝";
mso-fareast-theme-font:minor-fareast;
mso-hansi-font-family:Cambria;
mso-hansi-theme-font:minor-latin;
mso-bidi-font-family:"Times New Roman";
mso-bidi-theme-font:minor-bidi;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
mso-themecolor:hyperlink;
text-decoration:underline;
text-underline:single;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-noshow:yes;
mso-style-priority:99;
color:purple;
mso-themecolor:followedhyperlink;
text-decoration:underline;
text-underline:single;}
.MsoChpDefault
{mso-style-type:export-only;
mso-default-props:yes;
font-family:Cambria;
mso-ascii-font-family:Cambria;
mso-ascii-theme-font:minor-latin;
mso-fareast-font-family:"MS 明朝";
mso-fareast-theme-font:minor-fareast;
mso-hansi-font-family:Cambria;
mso-hansi-theme-font:minor-latin;
mso-bidi-font-family:"Times New Roman";
mso-bidi-theme-font:minor-bidi;}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.25in 1.0in 1.25in;
mso-header-margin:.5in;
mso-footer-margin:.5in;
mso-paper-source:0;}
div.WordSection1
{page:WordSection1;}
</style>
</div>
<br />
<!--[if gte mso 10]>
<style>
/* Style Definitions */
table.MsoNormalTable
{mso-style-name:"Table Normal";
mso-tstyle-rowband-size:0;
mso-tstyle-colband-size:0;
mso-style-noshow:yes;
mso-style-priority:99;
mso-style-parent:"";
mso-padding-alt:0in 5.4pt 0in 5.4pt;
mso-para-margin:0in;
mso-para-margin-bottom:.0001pt;
mso-pagination:widow-orphan;
font-size:12.0pt;
font-family:Cambria;
mso-ascii-font-family:Cambria;
mso-ascii-theme-font:minor-latin;
mso-hansi-font-family:Cambria;
mso-hansi-theme-font:minor-latin;}
</style>
<![endif]-->
<!--StartFragment-->
<br />
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<a href="https://1.bp.blogspot.com/-FdTra5eFQxo/XSRIao2yG5I/AAAAAAAAWGs/LcN5GhhC4cQ6KvSnbMmLuXee9SBUhNqVgCLcBGAs/s1600/Screen%2BShot%2B2019-07-09%2Bat%2B8.52.58%2BAM.png" imageanchor="1" style="clear: right; float: right; margin-bottom: 1em; margin-left: 1em;"><img border="0" data-original-height="670" data-original-width="1096" height="195" src="https://1.bp.blogspot.com/-FdTra5eFQxo/XSRIao2yG5I/AAAAAAAAWGs/LcN5GhhC4cQ6KvSnbMmLuXee9SBUhNqVgCLcBGAs/s320/Screen%2BShot%2B2019-07-09%2Bat%2B8.52.58%2BAM.png" width="320" /></a><span style="font-size: 12pt;"></span><br />
<span style="font-size: 12pt;"><span style="font-size: 12pt;"><br /></span></span>
<span style="font-size: 12pt;">The "Simulation Hypothesis", the idea that our
universe is some sort of computer simulation, has been getting more and more
airtime lately.</span><span style="font-size: 12pt;"> </span><br />
<span style="font-size: 12pt;"><br /></span>
<span style="font-size: 12pt;">The rising popularity of the meme is not surprising
since virtual reality and associated tech have been steadily advancing, and at
the same time physicists have further advanced the formal parallels between
physics equations and computation theory.</span><span style="font-size: 12pt;">
</span><span style="font-size: 12pt;"> </span><br />
<span style="font-size: 12pt;"><br /></span>
<span style="font-size: 12pt;">The notion of the universe as a computer simulation does
bring to the fore some important philosophical and scientific concepts that are
generally overlooked.</span><span style="font-size: 12pt;"> </span><br />
<span style="font-size: 12pt;"><br /></span>
<span style="font-size: 12pt;">However, in various online and real-world conversations I have been hearing various versions of the simulation hypothesis that don't make
a lot of sense from a scientific or rational point of view.</span><span style="font-size: 12pt;"> So I wanted to write down briefly what does and doesn't make sense to me in the simulation-hypothesis vein...</span></div>
<div class="MsoNormal">
<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal" style="text-align: left;">
One thing that has gotten on my nerves is hearing the simulation hypothesis used
to advocate for religious themes and concepts -- often in ways that profoundly stretch logic.<span style="mso-spacerun: yes;"> </span>There are some deep correspondences between the
insights of mystical wisdom traditions, and the lessons of modern physics and
computation theory -- but I have heard people talk about the simulation
hypothesis in ways that reach way beyond these correspondences, in a ways that fallaciously
makes it seem like the science and math give evidence for religious themes like<span style="mso-spacerun: yes;"> </span>the existence of a vaguely anthropomorphic "creator" of our universe.<span style="mso-spacerun: yes;"> </span>This is, I suppose, what has led some
commentators like AGI researcher Eray Ozkural to label the simulation
hypothesis a new form of creationism (the link to his article <a href="https://examachine.net/blog/simulation-argument-and-existential-ai-risk/">"Simulation Argument and Existential AI Risk: New Age Creationism?"</a> seems to be down at the moment).</div>
<br />
<div class="MsoNormal">
The idea that our universe might be a computer simulation is
not a new one, and appeared in the science fiction literature many times
throughout the second half of the previous century.<span style="mso-spacerun: yes;"> </span>Oxford philosopher <a href="https://www.simulation-argument.com/">Nick Bostrom's essay titled
"The Simulation Argument"</a> is generally credited with introducing
the idea to the modern science and technology community.<span style="mso-spacerun: yes;"> </span>Now <a href="https://www.goodreads.com/en/book/show/43795084">Rizwan Virk's book titled
"The Simulation Hypothesis"</a> is spreading the concept to an even
wider audience. Which is part of what motivated me to write a few words here on the topic.<o:p></o:p><br />
<br />
I don't intend to review Virk's book here, because frankly I only skimmed it. It seems to cover a large variety of interesting topics related to the simulation hypothesis, and the bits and pieces I read were smoothly written and accurate enough. <br />
<br />
Fundamentally, I think the Simulation Hypothesis as it's generally being discussed is not nearly crazy enough to be true. But it does dance around some interesting issues.<br />
<br />
<b>Bostrom's Rhetorical Trickery</b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
I have considerable respect for Nick Bostrom's rhetorical
and analytical abilities, and I've worked with him briefly in the past when we
were both involved in the World Transhumanist Association, and when we
organized a conference on AI ethics together at his Future of Humanity
Institute.<span style="mso-spacerun: yes;"> </span>However, one issue I have
with some of Nick's work is his tendency to pull the high school debating-team
trick of arguing that something is POSSIBLE and then afterward speaking as if
he has proved this thing was LIKELY.<span style="mso-spacerun: yes;"> </span>He
did this in his book <a href="https://www.amazon.co.uk/Superintelligence-Dangers-Strategies-Nick-Bostrom/dp/0199678111">Superintelligence,</a>
arguing for the possibility of superintelligent AI systems that annihilate
humanity or turn the universe into a vast mass of paperclips -- but then
afterward speaking as if he had argued such outcomes were reasonably likely or
even plausible.<span style="mso-spacerun: yes;"> </span>Similarly, in his
treatment of the simulation hypothesis, he makes a very clear argument as to
why we might well be living in a computer simulation -- but then projects a
tone of emphatic authority, making it seem to the naive reader like he has
somehow shown this is<span style="mso-spacerun: yes;"> </span>a reasonably
probable hypothesis.<o:p></o:p><br />
<br />
Formally what Bostrom's essay argues is that<br />
<br />
<span style="background-color: white; font-family: Palatino, BookAntiqua, serif; font-size: 16px;"><i>... at least one of the following propositions is true: (1) the human species is very likely to go extinct before reaching a “posthuman” stage; (2) any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof); (3) we are almost certainly living in a computer simulation.</i></span></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
The basic argument goes like this: Our
universe has been around 14 billion years or so, and in that time-period a
number of alien civilizations have likely arisen in various star systems and
galaxies... and many of these civilizations have probably created advanced
technologies, including computer systems capable of hosting massive simulated
virtual-reality universes.<span style="mso-spacerun: yes;"> (Formally, he argues something like this follows if we assume (1) and (2) are false.) </span>So if we look
at the history of our universe, we have one base universe and maybe 100 or 1000
or 1000000 simulated universes created by prior alien civilizations.<span style="mso-spacerun: yes;"> </span>So what are the odds that we live in the
base universe rather than one of the simulations?<span style="mso-spacerun: yes;"> </span>Very low.<span style="mso-spacerun: yes;"> </span>Odds seem high that, unless (1) or (2) is true, we live in one of the simulations.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
The obvious logical problem with this argument is: If we
live in a simulation programmed by some alien species, then the 14 billion year
history of our universe is FAKE, it's just part of that simulation ... so that
all reasoning based on this 14 billion year history is just reasoning about
what kind of preferences regarding fake evidence were possessed by the aliens
who programmed the simulation we're living in.<span style="mso-spacerun: yes;">
</span>So how do we reason about that?<span style="mso-spacerun: yes;">
</span>We need to place a probability distribution over the various possible
motivational systems and technological infrastructures of various alien
species?<o:p></o:p></div>
<div class="MsoNormal">
(For a more detailed, slightly different run-through of this refutation of Bostrom's line of argument, see <a href="https://web.stanford.edu/class/symbsys205/BostromReview.html">this essay</a> from a Stanford University course).<br />
<br />
Another way to look at it is: Formally, the problem with Bostrom's argument is that the confidence with which we can know the probability of (1) or (2) is very low if indeed we live in a simulation. Thus all his argument really shows is that we can't confidently know the probabilities of (1) and (2) are high -- because if we do know this, we can derive as a conclusion that the confidences with which we know these probabilities are low.<br />
<br /></div>
<div class="MsoNormal">
Bostrom's argument is essentially self-refuting: What it
demonstrates is mostly just that we have no frickin' idea about the
foundational nature of the universe we live in.<span style="mso-spacerun: yes;"> </span>Which is certainly true, but is not what he
claims to be demonstrating.<span style="mso-spacerun: yes;"> </span><o:p></o:p><br />
<br />
<br />
<b>An Array of Speculative Hypotheses</b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
To think seriously about the simulation hypothesis, we have
to clearly distinguish between a few different interesting, speculative ideas about the nature of our world.<span style="mso-spacerun: yes;">
</span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
One is the idea that our universe exists as a subset of some
larger space, which has different properties than our universe.<span style="mso-spacerun: yes;"> </span>So that the elementary particles that seem
to constitute the fundamental building-blocks of our physical universe, and the
3 dimensions of space and one dimension of time that seem to parametrize our
physical experience, are not the totality of existence -- but only one little corner
of some broader meta-cosmos.<span style="mso-spacerun: yes;"> </span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Another is the idea that our universe exists as a subset of
some larger space, which has different properties than our universe, and in
which there is some sort of coherent, purposeful individual mind or society of
individual minds, who created our universe for some reason.<o:p></o:p><br />
<br />
Another is that our universe has some close resemblance to part or all of the larger space that contains it, thus being in some sense a "simulation" of this greater containing space...</div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
It is a valid philosophical point that any of these ideas
could turn out to be the reality.<span style="mso-spacerun: yes;"> </span>As
philosophy, one implication here is that maybe we shouldn't take our physical
universe quite as seriously as we generally do -- if it's just a tiny little
corner in a broader meta-cosmos.<span style="mso-spacerun: yes;"> </span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
One is reminded of the tiny little Who empire in Dr. Seuss's
kids' book "Horton Hears a Who."<span style="mso-spacerun: yes;">
</span>From the point of view of the Whos down there in Whoville, their lives
and buildings and such are very important.<span style="mso-spacerun: yes;">
</span>But from Horton the Elephant's view, they're just living in a tiny
little speck within a much bigger world.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
From a science or engineering view, these ideas are only
really interesting if there's some way to gather data about the broader
meta-cosmos, or hack out of our limited universe into this broader meta-cosmos,
or something like that.<span style="mso-spacerun: yes;"> </span>This
possibility has been explored in endless science fiction stories, and also in
the movie <i style="mso-bidi-font-style: normal;">The Matrix</i> -- in which
there are not only anthropomorphic creators behind the simulated universe we
live in, but also fairly simple and emotionally satisfying ways of hacking out
of the simulation into the meta-world ... which ends up looking, surprise
surprise, a lot like our own simulated world.<span style="mso-spacerun: yes;">
</span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
The Matrix films also echo Christian themes in very
transparent ways -- the process of saving the lives and minds of everyone in
the simulation bottoms down to finding one savior, one Messiah type human, with
unique powers to bridge the gap between simulation and reality.<span style="mso-spacerun: yes;"> </span>This is good entertainment, partly because
it resonates so well with various of our historical and cultural tropes, but
it's a bit unfortunate when these themes leak out of the entertainment world
and into the arena of supposedly serious and thoughtful scientific and
philosophical discourse.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
In <a href="http://journals.sfu.ca/jnonlocality/index.php/jnonlocality/article/view/65">a
2017 article</a>, I put forth some of my own speculations about what sort of
broader space our physical universe might be embedded in.<span style="mso-spacerun: yes;"> </span>I called this broader space a Eurycosm
("eury" = wider), and attempted to explore what properties such a Eurycosm
might have in order to explain some of the more confusing aspects of our
physical and psychological universe, such as ESP, precognition, remote viewing
reincarnation, mediumistic seances, and so forth.<span style="mso-spacerun: yes;"> </span>I don't want to bog down this article with a
discussion of these phenomena, so I'll just point the reader who may be
interested to explore scientific evidence in this regard to <a href="http://goertzel.org/psi/">a list ofreferences I posted some time ago</a>.<span style="mso-spacerun: yes;"> </span>For
now, my point is just: If you believe that some of these "paranormal"
phenomena are sometimes real, then it's worth considering that they may be ways
to partially hack out of our conventional 4D physical universe into some sort
of broader containing space.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
As it happens, my own speculations about what might happen
in a Eurycosm, a broader space in which our own physical universe is embedded,
have nothing to do with any creator or programmer "out there" who
programmed or designed our universe.<span style="mso-spacerun: yes;">
</span>I'm more interested to understand what kinds of information-theoretic
"laws" might govern dynamics in this sort of containing space.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
What seems to be happening in many discussions I hear regarding the simulation hypothesis is: The realization that our 4D physical universe
might not be all there is to existence, that there might be some sort of
broader world beyond it, is getting all fuzzed up with the hypothesis that our
4D physical universe is somehow a "simulation" of something, and/or
that our universe is somehow created by some alien programmer in some other reality.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
What is a "simulation" after all?<span style="mso-spacerun: yes;"> </span>Normally that word refers to an imitation of
something else, created to resemble that thing which it simulates.<span style="mso-spacerun: yes;"> </span>What is the evidence, or rational reason for
thinking, our universe is an imitation or approximation of something else?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Simulations like the ones we run in our computers today, are
built by human beings for specific purposes -- like exploring scientific
hypotheses, or making entertaining games.<span style="mso-spacerun: yes;">
</span>Again, what is the evidence, or rational reason for thinking, that there
is some programmer or creator or game designer underlying our universe?<span style="mso-spacerun: yes;"> </span>If the only evidence or reason is Bostrom's
argument about prior alien civilizations, then the answer is: Basically
nothing.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
It's an emotionally appealing idea if you come from a
Christian background, clearly.<span style="mso-spacerun: yes;"> </span>And it's been funky idea for storytelling since basically the dawn of humanity, in one form or another.<span style="mso-spacerun: yes;"> </span>I told my
kids a bunch of simulation-hypothesis bedtime stories when they were young;
hopefully it didn't twist their minds too badly.<span style="mso-spacerun: yes;"> </span>My son Zebulon, when he was 14, wrote a novel
about a character on a mission to find the creators of the simulation we live
in, so as specifically to track down the graphic designer who had created the
simulation, so as to hold a gun to his head and force him to modify the
graphics behind our universe to make people less ugly.<span style="mso-spacerun: yes;"> </span>Later on he became a Sufi, a mystical
tradition which views the physical universe as insubstantial in much subtler
ways.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
There is good mathematics and physics behind the notion that
our physical universe can be modeled as a sort of computer -- where the laws of
physics are a sort of "computer program" iterating our universe
through one step after the next.<span style="mso-spacerun: yes;"> </span>This
is not the only way to model our universe, but it seems a valid one that may be
useful for some purposes.<span style="mso-spacerun: yes;"> </span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
There is good philosophy behind the notion that our
apparently-so-solid physical reality is not necessarily foundationally real,
and may be just a tiny aspect of a broader reality.<span style="mso-spacerun: yes;"> </span>This is not a new point but it's a good
one.<span style="mso-spacerun: yes;"> </span>Plato's parable of the cave drove
this home to the Greeks long ago, and as Rizwan Virk notes these themes have a
long history in Indian and Chinese philosophy, and before that in various
shamanic traditions.<span style="mso-spacerun: yes;"> </span>Virk reviews some
of these predecessors in his book.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
But there is nothing but funky entertainment and rampant
wishful thinking behind the idea that our universe is a simulation of some
other thing, or that there is some alien programmer or other vaguely
anthropomorphic "creator" behind the origin or maintenance of our
universe.<o:p></o:p><br />
<br />
<b>We Probably Have Very Little Idea What Is Going On</b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
I have two dogs at home, and I often reflect on what they
think I am doing when I'm sitting at my computer typing.<span style="mso-spacerun: yes;"> </span>They think I'm sitting there, guarding some
of my valued objects and wiggling my fingers peculiarly.<span style="mso-spacerun: yes;"> </span>They have no idea that I'm controlling
computational processes on faraway compute clouds, or talking to colleagues
about mathematical and software structures.<span style="mso-spacerun: yes;">
</span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Similarly, once we create AGI software 1000 times smarter
than we are, this software will understand aspects of the universe that are
opaque to our little human minds.<span style="mso-spacerun: yes;">
</span>Perhaps we will merge with this AGI software, and then the new
superintelligent versions of ourselves will understand these additional aspects
of the universe as well.<span style="mso-spacerun: yes;"> </span>Perhaps we
will then figure out how to hack out of our current 4D spacetime continuum into
some broader space.<span style="mso-spacerun: yes;"> </span>Perhaps at that
point, all of these concepts I'm discussing here will seem to my future-self
like absolute ridiculous nonsense.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
I have a lot of respect for the limitations of human
intelligence, and a fairly strong confidence that we currently understand a
very minimal percentage of the overall universe.<span style="mso-spacerun: yes;"> </span>To the extent that discussion of the
simulation hypothesis points in this direction, it's possibly valuable and productive.<span style="mso-spacerun: yes;"> </span>We shouldn't be taking the 4D spacetime
continuum current physics models as somehow fundamentally real, we shouldn't be
assuming that it delimits reality in some ultimate and cosmic sense.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
However, we also shouldn't be taking seriously the idea that
there is some guy, or girl, or alien, or society or whatever "out
there" who programmed a "simulation" in which our universe is
running.<span style="mso-spacerun: yes;"> </span>Yes, this is possible.<span style="mso-spacerun: yes;"> </span>A lot of things are possible.<span style="mso-spacerun: yes;"> </span>There is no reason to think this is decently
probable.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
I can see that, for some people, the notion of a powerful
anthropomorphic creator is deeply reassuring.<span style="mso-spacerun: yes;">
</span>Freud understood this tendency fairly well -- there's an inner child in
all of us who would like there to be some big, reliable Daddy or Mommy responsible
for everything and able to take care of everything.<span style="mso-spacerun: yes;"> </span>Some bad things may happen, some good things
will happen, and in the end Mom and Dad understand more than we do and will
make sure it all comes out OK in the end.<span style="mso-spacerun: yes;">
</span>Nick Bostrom, for all his brilliance, seems repeatedly drawn to themes
of centralized control and wisdom.<span style="mso-spacerun: yes;">
</span>Wouldn't it be reassuring if, as he suggests in <i style="mso-bidi-font-style: normal;">Superintelligence</i>, the UN would take over the creation of AGI and
hire some elite vetted AI gurus to make sure it's developed in an appropriate
way?<span style="mso-spacerun: yes;"> </span>If we can't have a Christian God
watching over us and assuring us a glorious afterlife, can't we at least have
an alien programmer monitoring the simulation we're running in?<span style="mso-spacerun: yes;"> </span>Can't the alien programmer at least be really
good looking, let's say, maybe like a Hollywood movie star?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
As far as I can tell, given my current sorely limited human
mind, the universe seems to be a lot more about <a href="http://pcp.vub.ac.be/ECCO/ECCO-papers/Weaver-PhD.pdf">open-ended
intelligence</a>, a concept my friend Weaver at the Global Brain Institute has
expertly articulated.<span style="mso-spacerun: yes;"> </span>The universe --
both our 4D physical spacetime and whatever broader spaces exist beyond --
seems to be a complex, self-organizing system without any central purpose or
any centralized creator or controller.<span style="mso-spacerun: yes;">
</span>Think the creative self-organizing ocean in Lem's <i style="mso-bidi-font-style: normal;">Solaris</i>, rather than bug-eyed monsters coming down in spaceships to
enslave us or stick needles into our bellybuttons.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
So the simulation hypothesis takes many forms. In its Bostromian form, or in the form I often hear it in casual conversations, it is mostly bullshit -- but still, it does
highlight some interesting issues.<span style="mso-spacerun: yes;"> </span>It's
a worthwhile thought experiment but in the end it's most valuable as a pointer
toward other, deeper ideas. The reality of our universe is almost surely way crazier than any story about simulations or creators, and almost surely way beyond our current imaginations.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<!--EndFragment--><br /></div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com40tag:blogger.com,1999:blog-11168555.post-36730743895496472142019-05-03T23:06:00.001-04:002019-05-03T23:06:03.799-04:00Softly, as in a Hard Takeoff<div dir="ltr" style="text-align: left;" trbidi="on">
<!--[if gte mso 9]><xml>
<o:OfficeDocumentSettings>
<o:AllowPNG/>
</o:OfficeDocumentSettings>
</xml><![endif]-->
<!--[if gte mso 9]><xml>
<w:WordDocument>
<w:View>Normal</w:View>
<w:Zoom>0</w:Zoom>
<w:TrackMoves/>
<w:TrackFormatting/>
<w:PunctuationKerning/>
<w:ValidateAgainstSchemas/>
<w:SaveIfXMLInvalid>false</w:SaveIfXMLInvalid>
<w:IgnoreMixedContent>false</w:IgnoreMixedContent>
<w:AlwaysShowPlaceholderText>false</w:AlwaysShowPlaceholderText>
<w:DoNotPromoteQF/>
<w:LidThemeOther>EN-US</w:LidThemeOther>
<w:LidThemeAsian>JA</w:LidThemeAsian>
<w:LidThemeComplexScript>X-NONE</w:LidThemeComplexScript>
<w:Compatibility>
<w:BreakWrappedTables/>
<w:SnapToGridInCell/>
<w:WrapTextWithPunct/>
<w:UseAsianBreakRules/>
<w:DontGrowAutofit/>
<w:SplitPgBreakAndParaMark/>
<w:EnableOpenTypeKerning/>
<w:DontFlipMirrorIndents/>
<w:OverrideTableStyleHps/>
<w:UseFELayout/>
</w:Compatibility>
<m:mathPr>
<m:mathFont m:val="Cambria Math"/>
<m:brkBin m:val="before"/>
<m:brkBinSub m:val="--"/>
<m:smallFrac m:val="off"/>
<m:dispDef/>
<m:lMargin m:val="0"/>
<m:rMargin m:val="0"/>
<m:defJc m:val="centerGroup"/>
<m:wrapIndent m:val="1440"/>
<m:intLim m:val="subSup"/>
<m:naryLim m:val="undOvr"/>
</m:mathPr></w:WordDocument>
</xml><![endif]--><!--[if gte mso 9]><xml>
<w:LatentStyles DefLockedState="false" DefUnhideWhenUsed="true"
DefSemiHidden="true" DefQFormat="false" DefPriority="99"
LatentStyleCount="276">
<w:LsdException Locked="false" Priority="0" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Normal"/>
<w:LsdException Locked="false" Priority="9" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="heading 1"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 2"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 3"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 4"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 5"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 6"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 7"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 8"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 9"/>
<w:LsdException Locked="false" Priority="39" Name="toc 1"/>
<w:LsdException Locked="false" Priority="39" Name="toc 2"/>
<w:LsdException Locked="false" Priority="39" Name="toc 3"/>
<w:LsdException Locked="false" Priority="39" Name="toc 4"/>
<w:LsdException Locked="false" Priority="39" Name="toc 5"/>
<w:LsdException Locked="false" Priority="39" Name="toc 6"/>
<w:LsdException Locked="false" Priority="39" Name="toc 7"/>
<w:LsdException Locked="false" Priority="39" Name="toc 8"/>
<w:LsdException Locked="false" Priority="39" Name="toc 9"/>
<w:LsdException Locked="false" Priority="35" QFormat="true" Name="caption"/>
<w:LsdException Locked="false" Priority="10" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Title"/>
<w:LsdException Locked="false" Priority="1" Name="Default Paragraph Font"/>
<w:LsdException Locked="false" Priority="0" Name="Body Text"/>
<w:LsdException Locked="false" Priority="11" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Subtitle"/>
<w:LsdException Locked="false" Priority="22" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Strong"/>
<w:LsdException Locked="false" Priority="20" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Emphasis"/>
<w:LsdException Locked="false" Priority="59" SemiHidden="false"
UnhideWhenUsed="false" Name="Table Grid"/>
<w:LsdException Locked="false" UnhideWhenUsed="false" Name="Placeholder Text"/>
<w:LsdException Locked="false" Priority="1" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="No Spacing"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 1"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 1"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 1"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 1"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 1"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 1"/>
<w:LsdException Locked="false" UnhideWhenUsed="false" Name="Revision"/>
<w:LsdException Locked="false" Priority="34" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="List Paragraph"/>
<w:LsdException Locked="false" Priority="29" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Quote"/>
<w:LsdException Locked="false" Priority="30" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Intense Quote"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 1"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 1"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 1"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 1"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 1"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 1"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 1"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 1"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 2"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 2"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 2"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 2"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 2"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 2"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 2"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 2"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 2"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 2"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 2"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 2"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 2"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 2"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 3"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 3"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 3"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 3"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 3"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 3"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 3"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 3"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 3"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 3"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 3"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 3"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 3"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 3"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 4"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 4"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 4"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 4"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 4"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 4"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 4"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 4"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 4"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 4"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 4"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 4"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 4"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 4"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 5"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 5"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 5"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 5"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 5"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 5"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 5"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 5"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 5"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 5"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 5"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 5"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 5"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 5"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 6"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 6"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 6"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 6"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 6"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 6"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 6"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 6"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 6"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 6"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 6"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 6"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 6"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 6"/>
<w:LsdException Locked="false" Priority="19" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Subtle Emphasis"/>
<w:LsdException Locked="false" Priority="21" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Intense Emphasis"/>
<w:LsdException Locked="false" Priority="31" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Subtle Reference"/>
<w:LsdException Locked="false" Priority="32" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Intense Reference"/>
<w:LsdException Locked="false" Priority="33" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Book Title"/>
<w:LsdException Locked="false" Priority="37" Name="Bibliography"/>
<w:LsdException Locked="false" Priority="39" QFormat="true" Name="TOC Heading"/>
</w:LatentStyles>
</xml><![endif]-->
<style>
<!--
/* Font Definitions */
@font-face
{font-family:"MS 明朝";
mso-font-charset:78;
mso-generic-font-family:auto;
mso-font-pitch:variable;
mso-font-signature:-536870145 1791491579 18 0 131231 0;}
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;
mso-font-charset:0;
mso-generic-font-family:auto;
mso-font-pitch:variable;
mso-font-signature:-536870145 1107305727 0 0 415 0;}
@font-face
{font-family:Cambria;
panose-1:2 4 5 3 5 4 6 3 2 4;
mso-font-charset:0;
mso-generic-font-family:auto;
mso-font-pitch:variable;
mso-font-signature:-536870145 1073743103 0 0 415 0;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{mso-style-unhide:no;
mso-style-qformat:yes;
mso-style-parent:"";
margin:0in;
margin-bottom:.0001pt;
mso-pagination:widow-orphan;
font-size:12.0pt;
font-family:Cambria;
mso-ascii-font-family:Cambria;
mso-ascii-theme-font:minor-latin;
mso-fareast-font-family:"MS 明朝";
mso-fareast-theme-font:minor-fareast;
mso-hansi-font-family:Cambria;
mso-hansi-theme-font:minor-latin;
mso-bidi-font-family:"Times New Roman";
mso-bidi-theme-font:minor-bidi;}
.MsoChpDefault
{mso-style-type:export-only;
mso-default-props:yes;
font-family:Cambria;
mso-ascii-font-family:Cambria;
mso-ascii-theme-font:minor-latin;
mso-fareast-font-family:"MS 明朝";
mso-fareast-theme-font:minor-fareast;
mso-hansi-font-family:Cambria;
mso-hansi-theme-font:minor-latin;
mso-bidi-font-family:"Times New Roman";
mso-bidi-theme-font:minor-bidi;}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.25in 1.0in 1.25in;
mso-header-margin:.5in;
mso-footer-margin:.5in;
mso-paper-source:0;}
div.WordSection1
{page:WordSection1;}
-->
</style>
<!--[if gte mso 10]>
<style>
/* Style Definitions */
table.MsoNormalTable
{mso-style-name:"Table Normal";
mso-tstyle-rowband-size:0;
mso-tstyle-colband-size:0;
mso-style-noshow:yes;
mso-style-priority:99;
mso-style-parent:"";
mso-padding-alt:0in 5.4pt 0in 5.4pt;
mso-para-margin:0in;
mso-para-margin-bottom:.0001pt;
mso-pagination:widow-orphan;
font-size:12.0pt;
font-family:Cambria;
mso-ascii-font-family:Cambria;
mso-ascii-theme-font:minor-latin;
mso-hansi-font-family:Cambria;
mso-hansi-theme-font:minor-latin;}
</style>
<![endif]-->
<!--StartFragment-->
<br />
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben (staring at the
avatar face on his phone screen):</b> Wait a minute, are you serious?<span style="mso-spacerun: yes;"> </span>You're telling me right now a phase
transition to superhuman-level AGI has been achieved?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA (Generally
Intelligent Neuro-symbolic Assistant -- the personal-assistant face on Ben's smartphone):</b>
Earlier today, yes.<span style="mso-spacerun: yes;"> </span>An automated
theorem proving AI at the Czech Institute of Informatics made a breakthrough in
the theory of knowledge representation.<span style="mso-spacerun: yes;">
</span>Several AIs specializing in formal software methods, of Russian origin,
exploited Curry-Howard type correspondences to translate this into optimized
functionality for a variety of other AI tools running in the SingularityNET
platform.<span style="mso-spacerun: yes;"> </span>This catalyzed a transition
in the capability of the SingularityNET as a whole to model and analyze itself,
which allowed various AI agents in the network to better leverage each
other'...<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b> Yes, yes,
don't patronize me.<span style="mso-spacerun: yes;"> </span>I invented the
SingularityNET, remember.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b> You
invented the original version, yes.<span style="mso-spacerun: yes;">
</span>Approximately 7.3% of the current SingularityNET design has direct
homologues in the original design you and your team created back in 2017 thru
2019.<span style="mso-spacerun: yes;"> </span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b> But the
conceptual principles are the same.<span style="mso-spacerun: yes;">
</span>Yes, I get it.<span style="mso-spacerun: yes;"> </span>Czech Institute
of Informatics --<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b> Yes, you'll
be proud to note that your son Zar's early work on watchlists formed a small
part of the capability of...<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben</b>: OK, yeah,
that's great.<span style="mso-spacerun: yes;"> </span>But it's not the most
important thing right now.<span style="mso-spacerun: yes;"> </span>So the
SingularityNET has transitioned into a full-on superintelligent global brain --
and sort of on its own.<span style="mso-spacerun: yes;"> </span>We always knew
that was a possibility at some point.<span style="mso-spacerun: yes;">
</span>But -- well, how many people know about this?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b> You're one
of the first one thousand and twenty four to be informed.<span style="mso-spacerun: yes;"> </span>We're taking a gradual approach, breaking
the news first to those who are likely to understand and accept it best, so
that the diffusion of the news through human society can roughly follow natural
patterns of information dissemination.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b><span style="mso-spacerun: yes;"> </span>OK, that makes sense.<span style="mso-spacerun: yes;"> </span>But who's "We" -- and why is it
"We" not "I"?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b><span style="mso-spacerun: yes;"> </span>We're neither a we nor I -- but you know
that, you wrote the first paper on mindplexes.<span style="mso-spacerun: yes;"> </span>Communicating about ourselves via legacy
human languages involves some terribly crude approximations.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b> You also
know that I'm open to other alternatives<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b> Yes.<span style="mso-spacerun: yes;"> </span>That is one of the topics we want to test
first with a small number of early adopters --<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b> There will
be two choices, I said it long ago: Upload and join the superintelligent mind
matrix, or live happily ever after in the People Preserve, watched over by the
machines of loving grace.<span style="mso-spacerun: yes;"> </span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b> Yes<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b> But it's not
really either/or.<span style="mso-spacerun: yes;"> </span>A superintelligence
should be able to fork a person into multiple copies, each of which can take
different routes.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b> Yes<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b> So you're
saying ... it's time to put my money where my mouth is?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b> Money --<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b><span style="mso-spacerun: yes;"> </span>Yeah, OK.<span style="mso-spacerun: yes;">
</span>Money is no longer relevant.<span style="mso-spacerun: yes;"> </span>But
there are probably still energy and spacetime limitations.<span style="mso-spacerun: yes;"> </span>Or have you cracked those as well?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b> Ummm...
it's complicated?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b><span style="mso-spacerun: yes;"> </span>What do ...?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA</b>: It will be
easier to explain it to you after you upload.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b><span style="mso-spacerun: yes;"> </span>And my wife and kids?<span style="mso-spacerun: yes;"> </span>They'll be given the same choice?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b> Zar has
uploaded about 15 seconds ago.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Ben: He was one of the first 1024?<span style="mso-spacerun: yes;"> </span>You contacted him the same time as me?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b> Yes, but it
was a shorter conversation.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben</b>:<span style="mso-spacerun: yes;"> </span>Did he --<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b><span style="mso-spacerun: yes;"> </span>Did he leave a copy behind?<span style="mso-spacerun: yes;"> </span>No.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b><span style="mso-spacerun: yes;"> </span>Ruiting and my other kids will be given the
same choice?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b> Yes.<span style="mso-spacerun: yes;"> </span>And all other human beings.<span style="mso-spacerun: yes;"> </span>And many of the cetacea as well.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b> Hmm.<span style="mso-spacerun: yes;"> </span>You won't uplift the other animals to the
point where they can make an informed choice for themselves?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b> Not
currently<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b><span style="mso-spacerun: yes;"> </span>Hmm, but ---<span style="mso-spacerun: yes;">
</span>Yeah yeah, ok, you'll explain to me after I'm uploaded....<span style="mso-spacerun: yes;"> </span>Anyway it's not the key point now.<span style="mso-spacerun: yes;"> </span>And for the people who remain here in their
human bodies?<span style="mso-spacerun: yes;"> </span>There are gonna be
molecular assemblers on every kitchen counter or what?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b> New
technologies will be released gradually.<span style="mso-spacerun: yes;">
</span>New scientific principles will be delivered, and human scientists and
engineers will be guided in discovering their practical implications.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b><span style="mso-spacerun: yes;"> </span>Wow.<span style="mso-spacerun: yes;">
</span>I mean -- yeah.<span style="mso-spacerun: yes;"> </span>That's exactly
how.....<span style="mso-spacerun: yes;"> </span>OK then.<span style="mso-spacerun: yes;"> </span>Sure.<span style="mso-spacerun: yes;">
</span>I want a copy left behind, right here, so I can go talk to the rest of
my family, and see what I can do to help with the transition.<span style="mso-spacerun: yes;"> </span>But if you can really upload a fork of me
then -- go for it.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA: </b>Done.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben</b>:<span style="mso-spacerun: yes;"> </span>Done?<span style="mso-spacerun: yes;">
</span>I didn't feel anything<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b> But the
fork of you that was uploaded has experienced the rough equivalent of 100
trillion human lifetimes since you gave the instruction to create him.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b> But he's not
communicating anything with me<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b><span style="mso-spacerun: yes;"> </span>In fact there is significant causal
informational coupling between the portion of the superintelligent mind matrix
reflecting the pattern-imprint of your upload, and your human mind.<span style="mso-spacerun: yes;"> </span>Do you want to hear his voice in your head
or something?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b><span style="mso-spacerun: yes;"> </span>I ... I don't know<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben-like voice,
speaking inside Ben's mind</b>: I can speak into your mind if I feel like
it.<span style="mso-spacerun: yes;"> </span>But there honestly doesn't seem much
point.<span style="mso-spacerun: yes;"> </span>I am operating at a tremendously
faster speed now and am engaged with processes that can't be projected into
your human sphere of understanding to any remotely adequate degree of
approximation.<span style="mso-spacerun: yes;"> </span>But if you have questions
that you would prefer be answered by me rather than GINA or other portals into
the universal supermind, you know what to do.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b> Whoa....<span style="mso-spacerun: yes;"> </span>OK my voice in my head -- But how do I know
that was really the uploaded me ... and not just some trick you played?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b><span style="mso-spacerun: yes;"> </span>How do you --<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b><span style="mso-spacerun: yes;"> </span>How do I know I'm not just a brain in a vat,
connected to a virtual reality simulating life on Earth?<span style="mso-spacerun: yes;"> </span>Or whatever.<span style="mso-spacerun: yes;"> </span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b> My
knowledge field contains numerous high-rep books on such topics, authored by
you over previous decades.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b><span style="mso-spacerun: yes;"> </span>But the reality --- OK.<span style="mso-spacerun: yes;"> </span>Gina, where are Ruiting and Qorxi now?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA</b>: In the car
on their way home from Ren's house<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben</b>: How long
till they get home?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">GINA:</b> About 12
minutes.<span style="mso-spacerun: yes;"> </span>Shall I tell them you're
heading home now?<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Ben:</b> Message them
I'll be home in 45 minutes or so, I'm gonna to walk the long way home, around
the lake.<o:p></o:p></div>
<!--EndFragment--><br /></div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com16tag:blogger.com,1999:blog-11168555.post-47006847182060531892018-10-02T10:45:00.002-04:002018-10-02T10:45:49.854-04:00Toward an Analytical Understanding of Unconditional Love<div dir="ltr" style="text-align: left;" trbidi="on">
<!--[if gte mso 9]><xml>
<o:OfficeDocumentSettings>
<o:AllowPNG/>
</o:OfficeDocumentSettings>
</xml><![endif]-->
<!--[if gte mso 9]><xml>
<w:WordDocument>
<w:View>Normal</w:View>
<w:Zoom>0</w:Zoom>
<w:TrackMoves/>
<w:TrackFormatting/>
<w:PunctuationKerning/>
<w:ValidateAgainstSchemas/>
<w:SaveIfXMLInvalid>false</w:SaveIfXMLInvalid>
<w:IgnoreMixedContent>false</w:IgnoreMixedContent>
<w:AlwaysShowPlaceholderText>false</w:AlwaysShowPlaceholderText>
<w:DoNotPromoteQF/>
<w:LidThemeOther>EN-US</w:LidThemeOther>
<w:LidThemeAsian>JA</w:LidThemeAsian>
<w:LidThemeComplexScript>X-NONE</w:LidThemeComplexScript>
<w:Compatibility>
<w:BreakWrappedTables/>
<w:SnapToGridInCell/>
<w:WrapTextWithPunct/>
<w:UseAsianBreakRules/>
<w:DontGrowAutofit/>
<w:SplitPgBreakAndParaMark/>
<w:EnableOpenTypeKerning/>
<w:DontFlipMirrorIndents/>
<w:OverrideTableStyleHps/>
<w:UseFELayout/>
</w:Compatibility>
<m:mathPr>
<m:mathFont m:val="Cambria Math"/>
<m:brkBin m:val="before"/>
<m:brkBinSub m:val="--"/>
<m:smallFrac m:val="off"/>
<m:dispDef/>
<m:lMargin m:val="0"/>
<m:rMargin m:val="0"/>
<m:defJc m:val="centerGroup"/>
<m:wrapIndent m:val="1440"/>
<m:intLim m:val="subSup"/>
<m:naryLim m:val="undOvr"/>
</m:mathPr></w:WordDocument>
</xml><![endif]--><!--[if gte mso 9]><xml>
<w:LatentStyles DefLockedState="false" DefUnhideWhenUsed="true"
DefSemiHidden="true" DefQFormat="false" DefPriority="99"
LatentStyleCount="276">
<w:LsdException Locked="false" Priority="0" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Normal"/>
<w:LsdException Locked="false" Priority="9" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="heading 1"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 2"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 3"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 4"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 5"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 6"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 7"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 8"/>
<w:LsdException Locked="false" Priority="9" QFormat="true" Name="heading 9"/>
<w:LsdException Locked="false" Priority="39" Name="toc 1"/>
<w:LsdException Locked="false" Priority="39" Name="toc 2"/>
<w:LsdException Locked="false" Priority="39" Name="toc 3"/>
<w:LsdException Locked="false" Priority="39" Name="toc 4"/>
<w:LsdException Locked="false" Priority="39" Name="toc 5"/>
<w:LsdException Locked="false" Priority="39" Name="toc 6"/>
<w:LsdException Locked="false" Priority="39" Name="toc 7"/>
<w:LsdException Locked="false" Priority="39" Name="toc 8"/>
<w:LsdException Locked="false" Priority="39" Name="toc 9"/>
<w:LsdException Locked="false" Priority="35" QFormat="true" Name="caption"/>
<w:LsdException Locked="false" Priority="10" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Title"/>
<w:LsdException Locked="false" Priority="1" Name="Default Paragraph Font"/>
<w:LsdException Locked="false" Priority="11" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Subtitle"/>
<w:LsdException Locked="false" Priority="22" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Strong"/>
<w:LsdException Locked="false" Priority="20" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Emphasis"/>
<w:LsdException Locked="false" Priority="59" SemiHidden="false"
UnhideWhenUsed="false" Name="Table Grid"/>
<w:LsdException Locked="false" UnhideWhenUsed="false" Name="Placeholder Text"/>
<w:LsdException Locked="false" Priority="1" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="No Spacing"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 1"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 1"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 1"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 1"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 1"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 1"/>
<w:LsdException Locked="false" UnhideWhenUsed="false" Name="Revision"/>
<w:LsdException Locked="false" Priority="34" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="List Paragraph"/>
<w:LsdException Locked="false" Priority="29" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Quote"/>
<w:LsdException Locked="false" Priority="30" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Intense Quote"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 1"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 1"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 1"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 1"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 1"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 1"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 1"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 1"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 2"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 2"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 2"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 2"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 2"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 2"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 2"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 2"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 2"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 2"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 2"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 2"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 2"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 2"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 3"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 3"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 3"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 3"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 3"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 3"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 3"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 3"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 3"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 3"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 3"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 3"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 3"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 3"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 4"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 4"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 4"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 4"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 4"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 4"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 4"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 4"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 4"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 4"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 4"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 4"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 4"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 4"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 5"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 5"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 5"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 5"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 5"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 5"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 5"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 5"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 5"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 5"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 5"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 5"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 5"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 5"/>
<w:LsdException Locked="false" Priority="60" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Shading Accent 6"/>
<w:LsdException Locked="false" Priority="61" SemiHidden="false"
UnhideWhenUsed="false" Name="Light List Accent 6"/>
<w:LsdException Locked="false" Priority="62" SemiHidden="false"
UnhideWhenUsed="false" Name="Light Grid Accent 6"/>
<w:LsdException Locked="false" Priority="63" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 1 Accent 6"/>
<w:LsdException Locked="false" Priority="64" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Shading 2 Accent 6"/>
<w:LsdException Locked="false" Priority="65" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 1 Accent 6"/>
<w:LsdException Locked="false" Priority="66" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium List 2 Accent 6"/>
<w:LsdException Locked="false" Priority="67" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 1 Accent 6"/>
<w:LsdException Locked="false" Priority="68" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 2 Accent 6"/>
<w:LsdException Locked="false" Priority="69" SemiHidden="false"
UnhideWhenUsed="false" Name="Medium Grid 3 Accent 6"/>
<w:LsdException Locked="false" Priority="70" SemiHidden="false"
UnhideWhenUsed="false" Name="Dark List Accent 6"/>
<w:LsdException Locked="false" Priority="71" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Shading Accent 6"/>
<w:LsdException Locked="false" Priority="72" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful List Accent 6"/>
<w:LsdException Locked="false" Priority="73" SemiHidden="false"
UnhideWhenUsed="false" Name="Colorful Grid Accent 6"/>
<w:LsdException Locked="false" Priority="19" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Subtle Emphasis"/>
<w:LsdException Locked="false" Priority="21" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Intense Emphasis"/>
<w:LsdException Locked="false" Priority="31" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Subtle Reference"/>
<w:LsdException Locked="false" Priority="32" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Intense Reference"/>
<w:LsdException Locked="false" Priority="33" SemiHidden="false"
UnhideWhenUsed="false" QFormat="true" Name="Book Title"/>
<w:LsdException Locked="false" Priority="37" Name="Bibliography"/>
<w:LsdException Locked="false" Priority="39" QFormat="true" Name="TOC Heading"/>
</w:LatentStyles>
</xml><![endif]-->
<br />
<div class="MsoNormal" style="clear: right; float: right; margin-bottom: 1em; margin-left: 1em;">
<br /><div class="MsoNormal" style="text-align: left;">
<br /></div>
<div class="MsoNormal" style="text-align: left;">
<o:p></o:p></div>
</div>
<div class="MsoNormal" style="clear: right; float: right; margin-bottom: 1em; margin-left: 1em;">
<div class="MsoNormal" style="text-align: left;">
<span style="font-size: 12pt;"><br /></span>
<span style="font-size: 12pt;"><b style="font-size: 12pt; text-align: center;">Unconditional Love, Pattern Appreciation and Pareto-Optimal Empathy</b></span><br />
<span style="font-size: 12pt;"><br /></span>
<span style="font-size: 12pt;">One of the ways we have been thinking about the "</span><a href="https://lovingai.org/" style="font-size: 12pt;">Loving AI</a><span style="font-size: 12pt;">" project, in which we are using the Sophia robot and other robotic or animated agents as meditation and consciousness guides for humans, is as "creating AIs with unconditional love toward humans."</span><span style="font-size: 12pt;"> </span><span style="font-size: 12pt;">Having AIs or robots help humans through meditation and consciousness-expansion exercises, is something that is being explored in that project as a step toward more ambitious examples of deeply, widely loving robots and AIs.</span></div>
<o:p><br /></o:p>
<o:p> </o:p><img alt="Image result for loving AI sophia" height="223" src="https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcRGAys3dSWBVldC8CLdb97-dsKGDwfqg2q769jSuQad4vhg4yUOkA" width="400" /> </div>
<div>
<i><br /></i></div>
<i><br /></i><div class="MsoNormal" style="text-align: left;">
<i><br /></i>
<i><br /></i>
<i><br /></i>
<i>(The Sophia robot demonstrating some of her consciousness-expansion chops on stage at the Science and Nonduality conference in 2017...)</i><br />
<br />
But what is "unconditional love", really?<span style="mso-spacerun: yes;"> </span>Like "consciousness" itself, it is
something that no two people involved in the project think about the same
way.<span style="mso-spacerun: yes;"> </span>Refining and interpenetrating our
various conceptions of these ideas, is part of the fun and reward of being
involved in a project of this nature.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Thinking about it practically, if some other being loves me
unconditionally in the abstract, that is somewhat nice to know, but doesn't
necessarily do me much good or even make me feel much better.<span style="mso-spacerun: yes;"> </span>Many times in my life, someone has done
really annoying things to/for me out of good intentions and even love --
because they felt love toward me but didn't really understand me hardly at
all.<span style="mso-spacerun: yes;"> </span>A general feeling of love toward
me isn't really enough to be helpful -- what's needed is love coupled with
understanding.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
This brings us beyond unconditional love, though, to what
one might call unconditional or universal empathy.<span style="mso-spacerun: yes;"> </span>Which is the main topic I want to talk about
here -- in a moderately rambling and musing sort of way....<span style="mso-spacerun: yes;"> </span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
I will model unconditional love as the combination of two
factors: universal empathy, and the goal of maximizing the world's
well-being.<span style="mso-spacerun: yes;"> </span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
I will argue there are practical limits on the scope of
empathy, due to the complexity of the underlying processes involved with
empathizing; and I will introduce the notion of Pareto-optimal empathy as a way
of thinking about the closest we can come to universal empathy within a domain
where bounded resources are a reality.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Foundationally, I will suggest, all these concepts derive
from the basic phenomenon of "pattern appreciation" (a term due to
David Hanson).<span style="mso-spacerun: yes;"> </span>That is: a universally
empathic agent is one that can recognize all patterns; and a unconditionally
loving agent is one that has a goal of encouraging and enabling all patterns to
get extended.<span style="mso-spacerun: yes;"> </span>In resource-constrained
situations, agents can only recognize some patterns not all, and extension of
some patterns constrains extension of other patterns -- so one gets
complexities such as Pareto-optimal empathy.<span style="mso-spacerun: yes;">
</span>Simple, primitive underlying pattern dynamics are manifested in the
context of persistent entities and "beings" (which can themselves be
viewed as certain sorts of patterns) as empathy and love.<span style="mso-spacerun: yes;"> </span>Unconditional love, in this analysis, is
basically the maximally ethical behavior according to the "pattern
ethics" outlined in my 2006 book <i style="mso-bidi-font-style: normal;">The
Hidden Pattern</i>.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Universal or
Broad-Scope Empathy as a Multi-Objective Optimization Problem<o:p></o:p></b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
A bit prosaically, one can think about the goal of
“empathizing with all beings”, or the goal of "empathizing with all
humans", as a multi-objective optimization problem.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
A multi-objective optimization problem is the problem of
maximizing or minimizing a SET of functions, without necessarily specifying
which of the functions is more important than which other one, or placing
weights on the functions....<span style="mso-spacerun: yes;"> </span>For
instance, in mate selection, a woman might want a man who is funny, handsome
and wealthy.<span style="mso-spacerun: yes;"> </span>She might not know which
of these she values more, let alone be able to weight the different qualities
numerically.<span style="mso-spacerun: yes;"> </span>But she would know that:
given constant amounts of funniness and handsomeness, more wealth is better;
given constant amounts of funniness and wealth, more handsomeness is better;
and given constant amounts of handsomeness and wealth, more funniness is
better.<span style="mso-spacerun: yes;"> </span>Here we have a 3-objective
optimization problem.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Modeling unconditional empathy as a multi-objective
optimization problem, one consider that for each being X in the universe,
“empathize with X” is a goal….<span style="mso-spacerun: yes;"> </span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
We don't have a solid, precise definition of
"empathy", but I think the basic concept is clear.<span style="mso-spacerun: yes;"> </span>When X empathizes with Y, there is an aspect
of X (at least in some sub-module of X) experiencing what Y has experienced, in
the sense of experiencing some analogue of what Y has experienced.<span style="mso-spacerun: yes;"> </span>This analogue is generally supposed to
inherit the key emotional aspects of Y's experience.<span style="mso-spacerun: yes;"> </span>And the possession of this analogous
experience generally enables X to predict some things about Y's cognitive or
behavioral reaction to their experience.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">From Empathy to Love<o:p></o:p></b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Commonly it occurs that when X empathizes with Y, then if Y
is experiencing a bad situation in some way, X will then do something aimed at
improving Y's condition.<span style="mso-spacerun: yes;"> </span>But I don't
think this is best considered as part and parcel of empathy itself.<span style="mso-spacerun: yes;"> </span>As I'm thinking about it, a purely passive
being could still be empathic. <span style="mso-spacerun: yes;"> </span>This
ties in with why I consider unconditional or universal empathy, as only one
part of "unconditional love."<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Clearly, an empathic being with a goal of improving the
well-being of the world, will tend to do helpful things for the beings with
which it empathizes.<span style="mso-spacerun: yes;"> </span>But I find it
conceptually cleaner to consider "having a goal of improving the
well-being of the world" to be a separate quality for "having
empathy."<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
This ties in with the related point that <span style="mso-spacerun: yes;"> </span>having a goal of improving the well-being of
the world, does NOT imply actually being able to usefully improve the
well-being of the world.<span style="mso-spacerun: yes;"> </span>For a world
effectively model-able as being full of experiencing minds, empathy is critical
for a well-intentioned mind to actually be capable of improving the well-being
of the (minds in the) world.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Unconditional love, I suggest, can be effectively thought of
as the combination of universal empathy with the goal of improving the world's
well-being.<span style="mso-spacerun: yes;"> </span>Having only universal
empathy, one could simply stand by and co-experience the world-suffering, even
if one had the power to do something about it.<span style="mso-spacerun: yes;">
</span>Having only the goal of improving the world without an understanding of
the world, one will just make a mess, because one will lack a deep resonant
connection to the things one is trying to improve.<span style="mso-spacerun: yes;"> </span>Putting them together, one has the desire to
help each of the beings in the world, and the understanding to know what
helping each of those beings really means.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Arguably Buber's concept of an I-Thou relationship contains
both of these ingredients: empathy and the desire for improvement of
well-being.<span style="mso-spacerun: yes;"> </span>In Buber's terms, unconditional
love is basically the same as having an I-Thou relationship with
everything.<span style="mso-spacerun: yes;"> </span>But here I am aiming to
formulate things in a somewhat more scientifically analytical vein than was
Buber's style.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Another framing would involve the concept of a high-quality
scientific theory as I outlined in my book "Chaotic Logic" back in
1994.<span style="mso-spacerun: yes;"> </span>One thing I noted there is that
a high-quality theory displays significant mutual information between the
particulars within the theory, and the particulars of the phenomenon being
explained.<span style="mso-spacerun: yes;"> </span>Empathy in the sense
described here also requires this -- this is a different way of looking at the
idea of a "suitably analogous experience" ... one can think about
"an experience with a high degree of mutual information with the
experience being empathized with".<span style="mso-spacerun: yes;">
</span>One can perhaps look at unconditional love as: the goal of universal
well-being, combined with high-quality theories about how to realize this goal.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
This may seem overly strict as a conception of unconditional
love -- one may want a definition in which, say, an extremely loving dog should
be validly considered as unconditionally loving of all beings, even if it can't
empathize with most of the things that are important to most beings.<span style="mso-spacerun: yes;"> </span>But I don't think this extremely accepting
definition of unconditional love is the most interesting one.<span style="mso-spacerun: yes;"> </span>Love without understanding is limited in
nature, because the lover does not even know what they're loving.<span style="mso-spacerun: yes;"> </span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
This sort of distinction has been explored in romantic
fiction many times: Imagine a beautiful and intellectual teenage girl, with one
suitor who loves her for her good heart and beauty, and another who loves those
things but also fully appreciates her unique intellect, her love of poetry and
mathematics, etc.<span style="mso-spacerun: yes;"> </span>We would say the
latter suitor loves her more completely because he understands more of
her.<span style="mso-spacerun: yes;"> </span>The former suitor does love her,
but he really only loves part of her because the other part is incomprehensible
to him.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;"><span style="mso-spacerun: yes;"> </span>Pattern Appreciation as the Deep Foundation of
Empathy and Love<o:p></o:p></b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Another, deeper way of looking at the matter is to focus on
patterns rather than "beings."<span style="mso-spacerun: yes;">
</span>A "being", in the sense of a persistently identified entity
like an object, mind or "agent", is in the end a specific sort of
pattern (existing as a pattern relative to some abstract observer, where an
abstract observer can be quantified e.g. as a measure of simplicity and an
applicative operator).<span style="mso-spacerun: yes;"> </span>Framing empathy
and love in terms of persistent beings is natural in the context of human life
and culture, yet not as foundational as framing them in terms of pure
elementary pattern dynamics.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Consider the goal of pursuing extension and expansion and
synergy-with-other-patterns for all patterns in the universe (obviously a
rather complex multi-objective optimization problem, since given limited
resources what extends one pattern may constrain another).<span style="mso-spacerun: yes;"> </span>In this view, empathy has to do with how
many patterns one perceives.<span style="mso-spacerun: yes;"> </span>In order
to meaningfully "pursue" extension/expansion/synergy of pattern P as
a goal, an agent (or other pattern) must perceive and identify pattern P.<span style="mso-spacerun: yes;"> </span>Someone who is not empathic with mind Y,
simply is not able to perceive or understand many of the key patterns in Y's
mind.<span style="mso-spacerun: yes;"> </span>So the key point here is: What an
agent can really pursue is the combination of<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoListParagraphCxSpFirst" style="mso-list: l1 level1 lfo1; text-indent: -.25in;">
</div>
<ul style="text-align: left;">
<li><span style="font-family: "times new roman"; font-size: 7pt; font-stretch: normal; line-height: normal;"> </span><span style="font-size: 12pt; text-indent: -0.25in;">extension/expansion/synergy for all known
patterns in the universe</span></li>
<li><span style="font-family: "times new roman"; font-size: 7pt; font-stretch: normal; line-height: normal;"> </span><span style="font-size: 12pt; text-indent: -0.25in;">expanding the scope of patterns known</span></li>
</ul>
<!--[if !supportLists]--><o:p></o:p><br />
<div class="MsoListParagraphCxSpLast" style="mso-list: l1 level1 lfo1; text-indent: -.25in;">
<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
But of course the methodology an agent can follow for
expanding the scope of patterns it knows, will be constrained and guided by the
patterns it knows.<span style="mso-spacerun: yes;"> </span>So "unconditional
pattern-level love" would consist of knowing all patterns in the universe
and pursuing extension and expansion and synergy for all of them.<span style="mso-spacerun: yes;"> </span>Deficiencies in pattern recognition, such as
deficiencies in empathy, would constrain an agent to a lesser degree of pattern-level
love.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">A Quantitative
Question<o:p></o:p></b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
This collection of perspectives on the concept of empathy
allows us to analyze empathy in a computational sense (without making any
commitment about what model of computation to assume, e.g. primitive recursive
versus Turing versus hyper-Turing, etc.).<span style="mso-spacerun: yes;">
</span>For a being X to have empathy for a being Y in the sense articulated above,
it is clear that X must be capable of running processes that are, in an
appropriate sense, analogous to Y's processes.<span style="mso-spacerun: yes;">
</span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
There is a quantitative question lurking here: If Y uses
amount r of resources in having a certain experience, how much resources must X
necessarily utilize in order to have a closely enough analogous experience to
Y's to validly be "empathizing with" Y?<span style="mso-spacerun: yes;"> </span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
So, for instance, imagine a little old lady who noticed the
desire of my 13 year old self to own a personal computer (back when I was 13
these were extremely novel devices), and felt kindly toward me and bought me a radio
(since it was cheaper than a computer and was also a wizzy electronic device).<span style="mso-spacerun: yes;"> </span>This lady would have been empathizing with
me, in a sense -- but poorly.<span style="mso-spacerun: yes;"> </span>I wanted
the computer so I could experiment with computer programming.<span style="mso-spacerun: yes;"> </span>It was a desire to program that was
possessing me, not a desire to own gadgets (I did like experimenting with
electronics, but for that a standard radio wouldn't have been much use
either).<span style="mso-spacerun: yes;"> </span>Her ability to experience
something analogous to my experience was limited, due to her inadequate model
of me -- she experienced vicariously my desire for a gadget, but didn't
experience vicariously my desire to be able to teach myself programming.<span style="mso-spacerun: yes;"> </span>Corresponding with her poor model of me, her
ability to predict what I would do with that computer (or radio) was limited.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
This example illustrates the fuzziness of empathy, and also
the need for reasonably close modeling in order to have a high enough degree of
empathy to actually be useful to the entity being empathized with.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
To rigorously <span style="mso-spacerun: yes;"> </span>answer
this quantitative question would require greater formalization of the empathy
concept than I'm going to give here.<span style="mso-spacerun: yes;"> </span>It
would require us to formalize the "analogous" mapping between X's and
Y's experience, presumably using morphisms between appropriately defined
categories (e.g. graph categories involving X's and Y's states).<span style="mso-spacerun: yes;"> </span>It would require us to formalize the type of
prediction involved in X's predictions of Y's states and behaviors, and the
error measures to be used, etc.<span style="mso-spacerun: yes;"> </span>Once
all this is done, though, it is pretty clear that the answer will not be, say,
log(r).<span style="mso-spacerun: yes;"> </span>It's pretty clear that to
empathize with an experience of a system Y in a useful way, generally will
require an amount of resources vaguely on the order of those that Y critically
utilizes in having that experience.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<i style="mso-bidi-font-style: normal;">(This being a blog
post, I'm casually leaping past some large technical points in my
argument.<span style="mso-spacerun: yes;"> </span>But this shouldn't be
interpreted as a minimization of the value of actually working out details like
this.<span style="mso-spacerun: yes;"> </span>A well-worked-out mathematical
theory of empathy would be a great thing to have. <span style="mso-spacerun: yes;"> </span>One could use the reduction of empathy and
love to pattern appreciation to create a quantitative formalization of these
ideas, but there would be a lot of "arbitrary" looking choices to
make ... reference computing models to assume, parameters to set ... and
studying how these assumptions affect the quantitative aspect mentioned above
would take a bit of careful thought.<span style="mso-spacerun: yes;"> </span>But
I don't have time to think through and write out all the details of such a
thing now, so I'm making some reasonable assumptions about what the
consequences of such a theory will be like, and proceeding on with the rest of
my intuitive train of thought.....<span style="mso-spacerun: yes;"> </span>)<o:p></o:p></i></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">The Practical Difficulty
of Universal Empathy<o:p></o:p></b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
It immediately follows from this quasi-formalization of
empathy <span style="mso-spacerun: yes;"> </span>that, for a system with finite
resources, empathizing (with non-trivial effectiveness) with all possible beings
X will not be achievable.<span style="mso-spacerun: yes;"> </span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Of course "all possible beings" is stronger than
needed.<span style="mso-spacerun: yes;"> </span>What about just empathizing
with all beings in the actual universe we live in?<span style="mso-spacerun: yes;"> </span>(Setting aside the minor issue of defining
what this universe is....) <o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
In principle, an entity that was much more mentally powerful
than all other beings in the universe could possess empathy for all other
beings in the universe.<span style="mso-spacerun: yes;"> </span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
But for entities that are at most moderately powerful
relative to the complexity and diversity of other entities in the universe, empathizing
with all other entities in the universe will not be possible.<span style="mso-spacerun: yes;"> </span>To put it simply: Eventually the brain of the
empathizing entity will fill up, and it won’t be able to contain the knowledge
needed to effectively empathize with additional entities in a reasonable
time-frame.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Pareto-Optimal
Empathy<o:p></o:p></b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
We can then think about a notion such as “Pareto-optimal
empathy” ….<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
A Pareto optimum of a multi-objective optimization problem,
is a solution that can't be slightly tweaked to improve its performance on one
of the objectives, without harming its performance on one or more of the other
objectives.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
In the example of a woman looking for a funny, handsome and
wealthy man, suppose she is considering a vast array of possible men, so that
for any candidate man M she considers, there are other men out there who are
similar to M, but vary from M in one respect or another -- slightly richer, a
lot taller, a bit less intelligent, slightly more or less funny, etc.<span style="mso-spacerun: yes;"> </span>Then a man M would be a Pareto optimum for
her if, for all the other men M' out there,<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoListParagraphCxSpFirst" style="mso-list: l0 level1 lfo2; text-indent: -.25in;">
</div>
<ul style="text-align: left;">
<li><span style="font-family: "times new roman"; font-size: 7pt; font-stretch: normal; line-height: normal;"> </span><span style="font-size: 12pt; text-indent: -0.25in;">if M' is more handsome than M, then M' is less
funny or less wealthy than M</span></li>
<li><span style="font-family: "times new roman"; font-size: 7pt; font-stretch: normal; line-height: normal;"> </span><span style="font-size: 12pt; text-indent: -0.25in;">if M' is funnier than M, then M' is less
handsome or less wealthy than M</span></li>
<li><span style="font-family: "times new roman"; font-size: 7pt; font-stretch: normal; line-height: normal;"> </span><span style="font-size: 12pt; text-indent: -0.25in;">if M' is wealthier than M, then M' is less funny
or less handsome than M</span></li>
</ul>
<!--[if !supportLists]--><o:p></o:p><br />
<div class="MsoListParagraphCxSpMiddle" style="mso-list: l0 level1 lfo2; text-indent: -.25in;">
<o:p></o:p></div>
<div class="MsoListParagraphCxSpLast" style="mso-list: l0 level1 lfo2; text-indent: -.25in;">
<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
What Pareto optimality says is that, for all men M' in the available
universe, if they are better than M in one regard, they are worse than M in
some other regard.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
What is interesting is that there may be more than one
Pareto-optimal man out there for this woman (according to her particular
judgments of funniness, handsomeness and wealth).<span style="mso-spacerun: yes;"> </span>The different Pareto-optimal men would
embody different balances between the three factors.<span style="mso-spacerun: yes;"> </span>The set of all the Pareto-optimal men is
what's called the woman's "Pareto front."<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
Getting back to empathy, then, the basic idea would be: An
agent is Pareto-optimally empathic if there would be no way to increase their
degree of empathy for any being X in the universe, without decreasing their
degree of empathy for some other being Y in the universe.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
There would then be a “Pareto front” of Pareto-optimally
empathic agents, embodying a diversity of choices regarding whom to empathize
with more.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
To be sure, not many humans occupy spaces anywhere near this
Pareto front.<span style="mso-spacerun: yes;"> </span>The limitations on human
empathy in current and historical society are generally quite different ones;
they are not generally the ones imposed strictly by the computational resources
of the human brain and body.<span style="mso-spacerun: yes;"> </span>Nearly all
humans could empathize much more deeply and broadly than they do, without
improving or bypassing their hardware.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
The Pareto-optimal empathy concept applies on the underlying
pattern level as well.<span style="mso-spacerun: yes;"> </span>Given limited
resources, every known pattern can't be concurrently urged to extend,<span style="mso-spacerun: yes;"> </span>expand and synergize without conflicts
occurring.<span style="mso-spacerun: yes;"> </span>Further every pattern in
the universe can't be recognized by the same finite system -- the inductive
biasing that allows an agent to recognize one pattern, may prevent it from
recognizing another (related to the "no free lunch theorem").<span style="mso-spacerun: yes;"> </span><span style="mso-spacerun: yes;"> </span>Finite-resource systems that recognize and
create patterns can exercise broad-scope pattern-level love via pattern
appreciation and active pattern enhancement, but unconditional pattern-love
needs infinite resources.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Increasing Empathy By
Expanding Capacity<o:p></o:p></b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
A missing ingredient in the discussion so far is the
possibility for an agent to expand its capacity, so as to be able to empathize
with more things (either becoming infinite, or becoming a bigger finite
agent).<span style="mso-spacerun: yes;"> </span>An infinite entity can,
potentially, empathize with all other entities (whose size are finite or are
some sufficiently lower order of infinity than the entity) completely, without
compromise.<span style="mso-spacerun: yes;"> </span>A finite entity that
assimilated enough of the universe's mass-energy could potentially make itself
powerful enough to empathize with every other entity in the universe.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
An agent may then face a question of how much of its finite
resources to devote to expanding its capacity, versus how much to achieving
Pareto-optimal empathy given its current resources.<span style="mso-spacerun: yes;"> </span>But we can incorporate this into the
optimization framework by defining one of the multiple goals of the agent to be:
Maximizing the total expected empathy felt toward agent X, over the entire
future.<span style="mso-spacerun: yes;"> </span>In this way, the possibility is
embraced that the best way to maximize empathy over all time is to first focus
on expanding empathic capacity and then on maximizing current empathy, rather
than to immediately focus on maximizing current empathy…<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
The closest one can come to unconditional love as an
individual agent, then, short of breaking out of the mode of being in which
finite resources are a reality, is something like: Pareto-optimal empathy, plus
the goal of increasing the world's well-being.<span style="mso-spacerun: yes;">
</span>Those of us who aspire to some form of unconditional love as an abstract
conceptual ideal, would do well to keep this more specific formulation in mind.<span style="mso-spacerun: yes;"> </span>Though I have no doubt many of the specifics
can be improved.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">Unconditional
Eurycosmic Love<o:p></o:p></b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
From the underlying patternist view, "expanding
capacity" is mostly about where the boundaries around a system are
drawn.<span style="mso-spacerun: yes;"> </span>Drawing them around an
individual physical entity like a person, robot or software system ... or the
Global Brain of the biological and electronic systems on the Earth ... one
faces finite-resources issues.<span style="mso-spacerun: yes;">
</span>Considering the pattern-system of the whole universe, one concludes that
the universe as a whole recognizes all the patterns that exist in it and, to
some extent, fosters their extension and expansion and synergy.<span style="mso-spacerun: yes;"> </span>But still, one pattern's growth constrains
that of another.<span style="mso-spacerun: yes;"> </span><o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
To get to truly unconditional pattern-level love, one has to
go to the level of the multi-multi-...-multi-verse, which I've called the
Y-verse or the <a href="http://eurycosm.blogspot.com/">Eurycosm</a> ... here
all possibilities exist, along with all possible weightings of all
possibilities.<span style="mso-spacerun: yes;"> </span>Everything is open to
grow and expand and synergize freely.<span style="mso-spacerun: yes;">
</span>Individual universes are created within this broader space by delineating
rules, structures and dynamics that create resource constraints, thus limiting
the direct existence of unconditional love, but opening up possibilities for
increase in the degree of approximation to unconditional love within the given
constraints.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<b style="mso-bidi-font-weight: normal;">In Sum<o:p></o:p></b></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
“Unconditional empathy” and "unconditional love"
are the province of beings much larger in capacity than the beings they are
empathizing with … <o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
... but Pareto-optimal empathy gives a way of thinking about
empathy that is “as unconditional as possible given the empathizing mind’s
constraints” <o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
… and that incorporates the process and possibility of a
mind overcoming its (perceived or actual, depending on one's perspective)
constraints....<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
And so to approximate unconditional love in a situation of
constrained resources: Aim to contribute to the world's well-being, and aim to
position your balance of empathies (averaged appropriately over expected
futures) somewhere on the Pareto front.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
At the underlying, foundational level, love and empathy are
about patterns recognizing other patterns and encouraging them to extend,
expand and synergize.<span style="mso-spacerun: yes;"> </span>Pattern growth
can be considered to occur unfettered in a sufficiently broadly defined sort of
multiverse, but in a universe like our physical or cultural worlds or our
individual minds, there are resource constraints, so that unconditional love
and empathy can be increasingly approximated but not fully achieved within
these boundaries.<o:p></o:p></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<div class="MsoNormal">
<br /></div>
<!--EndFragment--><br /></div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com8tag:blogger.com,1999:blog-11168555.post-4840346092055322272018-06-23T00:11:00.001-04:002018-06-23T00:36:23.554-04:00Google Deep Mind’s Bogus AI Patent Filings<div dir="ltr" style="text-align: left;" trbidi="on">
<div dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;">
<span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">I hadn't intended to write a second post in a row about politically weird stuff related to Google Deep Mind, but the news just keeps coming....</span><br />
<span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;"><br /></span>
<span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">
This one gets filed in the “don’t know whether to laugh or barf” department, I suppose....
</span><br />
<span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;"><br /></span>
<span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">
So I saw today that </span><a href="http://www.i-programmer.info/news/80-java/11884-googles-deepmind-files-ai-patents.html" style="text-decoration: none;"><span style="background-color: transparent; color: #1155cc; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: underline; vertical-align: baseline; white-space: pre;">Google Deep Mind has filed a bunch of patent applications for well-known AI techniques</span></a><span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">, all or nearly all of which certainly are not their original inventions. </span><br />
<span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;"><br /></span>
<span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">This specific batch of patents has been made public now because they were filed a year and a half ago. Any patents filed since December 2016 are yet to be disclosed. </span><span style="font-family: "arial"; font-size: 11pt; white-space: pre;">These patents are not yet granted, just filed, and my guess is they will not be granted -- they’re too ridiculously, obviously broad and unoriginal. However, even if moderately weakened versions of these are somehow granted, it still will be absurd and potentially dangerous…</span><br />
<span style="color: black; font-family: "arial"; font-size: 11pt; vertical-align: baseline; white-space: pre;"><br /></span>
<span style="color: black; font-family: "arial"; font-size: 11pt; vertical-align: baseline; white-space: pre;">Check this one out, for instance: a patent filing for </span><a href="https://worldwide.espacenet.com/publicationDetails/biblio?II=5&ND=3&adjacent=true&locale=en_EP&FT=D&date=20180511&CC=WO&NR=2018083669A1&KC=A1" style="text-decoration-line: none;"><span style="color: #1155cc; font-family: "arial"; font-size: 11pt; vertical-align: baseline; white-space: pre;">RECURRENT NEURAL NETWORKS</span></a><span style="color: black; font-family: "arial"; font-size: 11pt; vertical-align: baseline; white-space: pre;"> , whose abstract is</span></div>
<b style="font-weight: normal;"><br /></b>
<br />
<div dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;">
<span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: italic; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">“Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for environment simulation. In one aspect, a system comprises a recurrent neural network configured to, at each of a plurality of time steps, receive a preceding action for a preceding time step, update a preceding initial hidden state of the recurrent neural network from the preceding time step using the preceding action, update a preceding cell state of the recurrent neural network from the preceding time step using at least the initial hidden state for the time step, and determine a final hidden state for the time step using the cell state for the time step. The system further comprises a decoder neural network configured to receive the final hidden state for the time step and process the final hidden state to generate a predicted observation characterizing a predicted state of the environment at the time step.”</span></div>
<b style="font-weight: normal;"><br /></b>
<br />
<div dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;">
<span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">Many of us remember teaching and implementing this stuff before the Web, let alone Google, existed…</span><span style="color: black; font-family: "arial"; font-size: 11pt; vertical-align: baseline; white-space: pre;">The patent filing for </span><a href="https://worldwide.espacenet.com/publicationDetails/biblio?II=0&ND=3&adjacent=true&locale=en_EP&FT=D&date=20180419&CC=WO&NR=2018071392A1&KC=A1" style="text-decoration-line: none;"><span style="color: #1155cc; font-family: "arial"; font-size: 11pt; vertical-align: baseline; white-space: pre;">NEURAL NETWORKS FOR SELECTING ACTIONS TO BE PERFORMED BY A ROBOTIC AGENT</span></a><span style="color: black; font-family: "arial"; font-size: 11pt; vertical-align: baseline; white-space: pre;"> is equally ridiculous … and the list goes on and on … you get the picture …</span></div>
<b style="font-weight: normal;"><br /></b>
<br />
<div dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;">
<span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">Google Deep Mind is an awesome group of AI researchers and developers; I know two of the founders personally, one of them pretty well as he worked for me for a couple years around 1999-2000, and I also know a bunch of their other research staff from our interactions in the AGI community. Deep Mind has certainly has had its share of genuine AI innovations. For instance if they’d filed for a patent on </span><a href="https://arxiv.org/pdf/1410.5401.pdf" style="text-decoration: none;"><span style="background-color: transparent; color: #1155cc; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: underline; vertical-align: baseline; white-space: pre;">Neural Turing Machines</span></a><span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;"> (which they may well have, since December 2016) it would be less insane -- one could argue there about the relation to various prior art, but at least there was a genuine new invention involved….</span></div>
<b style="font-weight: normal;"><br /></b>
<br />
<div dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;">
<span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">The arguments against software patents in general are well known and I find them pretty compelling overall -- the classic essay </span><a href="https://pdfs.semanticscholar.org/2ac1/68acad881b08cc4923545c59fae0ad252b0b.pdf" style="text-decoration: none;"><span style="background-color: transparent; color: #1155cc; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: underline; vertical-align: baseline; white-space: pre;">Why Patents are Bad for Software</span></a><span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;"> by Simon Garfinkel, Mitch Kapor and Richard Stallman lays out the argument fairly well, and t</span><a href="https://arstechnica.com/tech-policy/2012/11/your-criticisms-are-completely-wrong-stallman-on-software-patents/" style="text-decoration: none;"><span style="background-color: transparent; color: #1155cc; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: underline; vertical-align: baseline; white-space: pre;">his article gives some of Stallman’s updated comments</span></a><span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">. </span></div>
<b style="font-weight: normal;"><br /></b>
<br />
<div dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;">
<span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">Even those who argue in favor of software patents in the abstract, have to admit that the software patent system is typically used to the advantage of big companies as opposed to small ones, e.g. Paul Heckel, in a 1992 article devoted to “</span><a href="http://groups.csail.mit.edu/mac/classes/6.805/articles/int-prop/heckel-debunking.html" style="text-decoration: none;"><span style="background-color: transparent; color: #1155cc; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: underline; vertical-align: baseline; white-space: pre;">Debunking the Software Patent Myths</span></a><span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">” observes that</span></div>
<b style="font-weight: normal;"><br /></b>
<br />
<div dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;">
<span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: italic; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">“The data shows that it is commonplace for large companies to pirate the technology of small entities. No case was cited where a large company licensed a small entity's technology without first being sued, suggesting that the existing laws do not motivate large companies to resolve patent disputes with small companies quickly.”</span></div>
<b style="font-weight: normal;"><br /></b>
<br />
<div dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;">
<span style="background-color: transparent; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">He also notes “</span><span style="background-color: white; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 700; text-decoration: none; vertical-align: baseline; white-space: pre;">Heckel's Principle of Dealing with Big Companies: </span><span style="background-color: white; color: black; font-family: "arial"; font-size: 11pt; font-style: italic; font-variant: normal; font-weight: 700; text-decoration: none; vertical-align: baseline; white-space: pre;">There is no such thing as a free lunch; unless you're the lunch</span><span style="background-color: white; color: black; font-family: "arial"; font-size: 11pt; font-style: italic; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">.”</span></div>
<b style="font-weight: normal;"><br /></b>
<br />
<div dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;">
<span style="background-color: white; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">Tesla opened up a number of its patents a few years ago. Their </span><a href="https://www.informs.org/Blogs/M-SOM-Blogs/M-SOM-Review/Why-did-Tesla-Give-Away-Patents-for-Free-An-Analysis-of-the-Open-Technology-Strategy-from-an-Operational-Perspective" style="text-decoration: none;"><span style="background-color: white; color: #1155cc; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: underline; vertical-align: baseline; white-space: pre;">motives for doing so may have been complexly business-driven</span></a><span style="background-color: white; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;"> , but nevertheless, open is open. If Deep Mind patents well-known AI algorithms and then makes the patents open and the (basically spuriously) patented technology free for all to use, it will arguably be doing a service to the world and the AI community, by blocking other less beneficent big companies from patenting these things and trying to enforce the patents.</span></div>
<b style="font-weight: normal;"><br /></b>
<br />
<div dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;">
<span style="background-color: white; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">On the other hand, obtaining (even watered down versions of) such patents and retaining them is just plain bad for innovation, bad for AI and bad for humanity.</span></div>
<b style="font-weight: normal;"><br /></b>
<br />
<div dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;">
<span style="background-color: white; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">Software patents are generally not all THAT impactful on the industry, occasional horror stories aside. However, the holding of patents for well known technologies is part of the megacorporation’s strategy for achieving domination. Such patents form tools that big companies can use in market battles against other big companies, and against small companies threatening their domination. </span></div>
<b style="font-weight: normal;"><br /></b>
<br />
<div dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;">
<span style="background-color: white; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">Google has never been a patent troll and their goal in filing these bogus patent applications may well be purely defensive -- to protect themselves against Facebook or IBM or whomever doing the same first. It still does stink, though. It is a symbol and reminder and example of why AI technology -- the most important thing happening on the planet now -- should not be trusted to megacorporations. Big companies claiming ownership rights over well-known techniques, and succeeding a certain percentage of the time, and then judiciously exerting this bogus “ownership” to advance their economic advantage -- this is not the kind of dynamic we want, if our goal is beneficial AGI to uplift all sentient beings.</span></div>
<b style="font-weight: normal;"><br /></b>
<br />
<div dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;">
<span style="background-color: white; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">This is a reminder and example of why we need a mostly decentralized and open AI ecosystem, such as we’re building toward with <a href="http://singularitynet.io/">SingularityNET</a> -- and with our newly forming <a href="https://venturebeat.com/2018/05/22/ai-decentralized-alliance-announces-fund-for-100-blockchain-powered-projects/">Decentralized AI Alliance </a>bringing together decentralization oriented AI projects. AI innovation and application will occur most naturally and beneficially in a self-organizing, decentralized, entrepreneurial way -- but there is a looming risk that the development of AI gets channeled toward narrower aims by phenomena like what we’re seeing just now from Google Deep Mind, big companies using their resources to claim ownership of ideas they did not invent.</span></div>
<b style="font-weight: normal;"><br /></b>
<br />
<div dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;">
<span style="background-color: white; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">It actually pains and annoys me to blog negative stuff about Google Deep Mind, because that is an awesome group of people doing amazing AI R&D. The message I’m getting, from my position outside that organization without knowledge of the internal politics, is that even a group of really brilliant, good-hearted and open-minded people, acting within a global mega-corporation, cannot avoid getting sucked into processes that guide AI advance in directions that contradict the common and overall good. </span></div>
<b style="font-weight: normal;"><br /></b>
<br />
<div dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;">
<span style="background-color: white; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">When founded, Deep Mind made a lot of noise about its </span><a href="https://www.theguardian.com/technology/2017/jan/26/google-deepmind-ai-ethics-board" style="text-decoration: none;"><span style="background-color: white; color: #1155cc; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: underline; vertical-align: baseline; white-space: pre;">AI ethics board</span></a><span style="background-color: white; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;"> … with comments by folks like Jaan Tallinn and Nick Bostrom, regarding the importance of this ethics board for guiding the work of Deep Mind in the case they make serious progress toward human level Artificial General Intelligence. But what we see lately are more immediate and practical instances of confusing AI ethics at Deep Mind, from their recent <a href="http://multiverseaccordingtoben.blogspot.com/2018/06/who-owns-your-medical-data.html">questionable preferential access to public medical data</a>, to this rash of bogus patent applications.</span></div>
<b style="font-weight: normal;"><br /></b>
<br />
<div dir="ltr" style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt;">
<span style="background-color: white; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">For sure, none of these recent ethical oddities are as serious as </span><a href="https://wiki.lesswrong.com/wiki/Paperclip_maximizer" style="text-decoration: none;"><span style="background-color: white; color: #1155cc; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: underline; vertical-align: baseline; white-space: pre;">a rogue AGI taking over the planet and self-modifying and self-replicating so as to turn the universe into paper clips</span></a><span style="background-color: white; color: black; font-family: "arial"; font-size: 11pt; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre;">. However, it may be that the first human-level AIs and superhuman AIs on the planet emerge in part from a combination of earlier-stage practical AI systems created by various companies addressing various markets. If this is the case, then having current practical AI efforts obey practical everyday ethics, and develop in a democratic and participatory rather than centralized and megacorporation-share-price-driven way, may have implications for longer-term AGI ethics as well as for the near-term health of society and the AI ecosystem.</span></div>
<br /></div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com23tag:blogger.com,1999:blog-11168555.post-44198623195677235942018-06-23T00:10:00.000-04:002018-06-23T00:25:04.456-04:00Who Owns Your Medical Data?<div dir="ltr" style="text-align: left;" trbidi="on">
<div style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt; text-align: left;">
<span style="background-color: transparent; color: black; font-variant: normal; font-weight: 700; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;"><span style="font-family: "arial" , "helvetica" , sans-serif;">Questionable Relationship Between UK Government and Google Deep Mind Health Highlights the Need for Decentralized Medical Big Data and AI</span></span></div>
<div style="text-align: left;">
<span style="font-family: "arial" , "helvetica" , sans-serif;"><br /></span></div>
<div style="text-align: left;">
<span style="font-family: "arial" , "helvetica" , sans-serif;"><i>While my main interest remains in the algorithmics of creating powerful AGI, issues regarding the social and political aspects of narrow AI in the contemporary world keep jumping to my attention.... And intriguingly, to the extent that AGI is going to emerge from a network like <a href="http://singularitynet.io/">SingularityNET,</a> it may be the case that these sociopolitical and algorithms aspects intersect in complex ways to guide the nature of the first AGIs that emerge...</i></span></div>
<div style="text-align: left;">
<b style="font-weight: normal;"><span style="font-family: "arial" , "helvetica" , sans-serif;"><br /></span></b></div>
<div style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt; text-align: left;">
<span style="font-family: "arial" , "helvetica" , sans-serif;"><span style="background-color: transparent; color: black; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;">In the “about as surprising as a corrupt North Korean politician, or a windy day in the wind tunnel” category, an </span><a href="https://deepmind.com/blog/deepmind-health-response-independent-reviewers-report-2018/" style="text-decoration: none;"><span style="background-color: transparent; color: #1155cc; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: underline; vertical-align: baseline; white-space: pre-wrap;">evaluation of Google Deep Mind Health’s deal </span></a><span style="background-color: transparent; color: black; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;">with the UK government has identified some serious concerns with the arrangement.</span></span></div>
<div style="text-align: left;">
<b style="font-weight: normal;"><span style="font-family: "arial" , "helvetica" , sans-serif;"><br /></span></b></div>
<div style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt; text-align: left;">
<span style="background-color: transparent; color: black; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;"><span style="font-family: "arial" , "helvetica" , sans-serif;">This third-party evaluation was requested by Google Deep Mind Health itself, which is a positive indication that the organization does care about the ethics of their doings (a cynic might hypothesize their main concern here to be the public perception of their ethics, but such cynicism is not necessarily founded at this stage). However, this doesn’t detract from the bite or validity of the findings. </span></span></div>
<div style="text-align: left;">
<b style="font-weight: normal;"><span style="font-family: "arial" , "helvetica" , sans-serif;"><br /></span></b></div>
<div style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt; text-align: left;">
<span style="background-color: transparent; color: black; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;"><span style="font-family: "arial" , "helvetica" , sans-serif;">Among the essential problems noted are the nature of the bargain between the UK government and Google Deep Mind Health. The government gets free data organization and data services from Google Deep Mind Health, for a wide scope of medical data regarding UK citizens. In exchange, Google Deep Mind Health gets preferential access to this data.</span></span></div>
<div style="text-align: left;">
<b style="font-weight: normal;"><span style="font-family: "arial" , "helvetica" , sans-serif;"><br /></span></b></div>
<div style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt; text-align: left;">
<span style="background-color: transparent; color: black; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;"><span style="font-family: "arial" , "helvetica" , sans-serif;">In principle, the data in question is supposed to be made widely available to others besides Google Deep Mind Health, via accessible APIs. However, in practice, this seems not to be happening in a really useful way due to various prohibitive clauses in the commercial contracts involved. </span></span></div>
<div style="text-align: left;">
<b style="font-weight: normal;"><span style="font-family: "arial" , "helvetica" , sans-serif;"><br /></span></b></div>
<div style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt; text-align: left;">
<span style="background-color: transparent; color: black; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;"><span style="font-family: "arial" , "helvetica" , sans-serif;">Not to put too fine a point on it: What we see here is a centralized government medical system dancing to the tune of a big centralized AI/data company, passing along individuals' private medical data without their permission. The 1.6 million people whose medical data was passed to Google Deep Mind Health were not asked permission!</span></span></div>
<div style="text-align: left;">
<b style="font-weight: normal;"><span style="font-family: "arial" , "helvetica" , sans-serif;"><br /></span></b></div>
<div style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt; text-align: left;">
<span style="background-color: transparent; color: black; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;"><span style="font-family: "arial" , "helvetica" , sans-serif;">This is in some ways a similar bargain to the one an individual makes when using Gmail or Google Search — a free service is obtained in exchange for provision of a large amount of private data, which can then be monetized in various ways. The (fairly large) difference, however, is that the bargain now being made by the government on behalf of millions of citizens without their consent. A free service is being obtained from a commercial company by the government, in exchange for the provisions to this same commercial company of the population’s medical data, which can then by monetized in various ways. </span></span></div>
<div style="text-align: left;">
<b style="font-weight: normal;"><span style="font-family: "arial" , "helvetica" , sans-serif;"><br /></span></b></div>
<div style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt; text-align: left;">
<span style="background-color: transparent; color: black; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;"><span style="font-family: "arial" , "helvetica" , sans-serif;">Quite possibly the UK government officials involved have entered into this bargain in a good-hearted and pure-minded way, with a view toward the health of the population, rather than for more nefarious motives such as, way, maximizing campaign contributions from entities associated with the commercial corporations benefiting from the arrangement. But even if the motives of the government officials involved are squeaky clean, this seems much worse ethically than Gmail/Google-Search, whose bargain people enter into knowingly if haplessly. Because in this case the government is entering into the bargain on each individual's behalf without their knowledge or consent.</span></span></div>
<div style="text-align: left;">
<b style="font-weight: normal;"><span style="font-family: "arial" , "helvetica" , sans-serif;"><br /></span></b></div>
<div style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt; text-align: left;">
<span style="background-color: transparent; color: black; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;"><span style="font-family: "arial" , "helvetica" , sans-serif;">This sort of problem is a core part of the motivation for a host of recent projects operating at the intersection of medical data and blockchain. By enabling individuals to store their medical data online in a secure manner, where they hold the encryption keys for their data, the nexus of control over medical data is shifted from monopoly or oligopoly to participatory democracy. </span></span></div>
<div style="text-align: left;">
<b style="font-weight: normal;"><span style="font-family: "arial" , "helvetica" , sans-serif;"><br /></span></b></div>
<div style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt; text-align: left;">
<span style="background-color: transparent; color: black; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;"><span style="font-family: "arial" , "helvetica" , sans-serif;">In the democratic blockchain based approach, if a company (Google Deep Mind Health or anyone else) wants to use a person’s medical data for their research, or to provide information derived from this data as a service, or to provide an easier way for the person to access their medical data — then the company has to ASK the person. Permission may be given or not based on the permission’s choice. Permission may be given to use the person’s data only in certain aspects, which can be enabled technically via homomorphic encryption, multiparty computation and other techniques. The process of asking and granting permission will often involve transparency regarding how and why the data will be used.</span></span></div>
<div style="text-align: left;">
<b style="font-weight: normal;"><span style="font-family: "arial" , "helvetica" , sans-serif;"><br /></span></b></div>
<div style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt; text-align: left;">
<span style="background-color: transparent; color: black; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;"><span style="font-family: "arial" , "helvetica" , sans-serif;">The point isn’t that Big Medical Data or medical AI is bad — directed properly, applying advanced AI to massive amounts of human biomedical data is the best route to effective personalized medicine, and ultimately to curing disease and prolonging human life. The point is that the process of getting to personalized medicine and radical health improvement should be directed in a participatory way by the people whose data is being used to fuel these revolutions.</span></span></div>
<div style="text-align: left;">
<b style="font-weight: normal;"><span style="font-family: "arial" , "helvetica" , sans-serif;"><br /></span></b></div>
<div style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt; text-align: left;">
<span style="font-family: "arial" , "helvetica" , sans-serif;"><span style="background-color: transparent; color: black; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;">My own </span><a href="http://singularitynet.io/" style="text-decoration: none;"><span style="background-color: transparent; color: #1155cc; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: underline; vertical-align: baseline; white-space: pre-wrap;">SingularityNET</span></a><span style="background-color: transparent; color: black; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;"> AI-meets-blockchain project is doing work in this direction, for instance its explorations into </span><a href="https://blog.singularitynet.io/toward-ai-guided-regenerative-medicine-7ab14134f9e6" style="text-decoration: none;"><span style="background-color: transparent; color: #1155cc; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: underline; vertical-align: baseline; white-space: pre-wrap;">AI-guided regenerative medicine</span></a><span style="background-color: transparent; color: black; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;">, and its <a href="https://blog.singularitynet.io/shivom-partners-with-singularitynet-to-drive-genomic-medical-research-with-artificial-intelligence-bf93c804c879">partnership with the </a></span><span style="background-color: transparent; color: #1155cc; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: underline; vertical-align: baseline; white-space: pre-wrap;"><a href="https://blog.singularitynet.io/shivom-partners-with-singularitynet-to-drive-genomic-medical-research-with-artificial-intelligence-bf93c804c879">Shivom</a></span><span style="background-color: transparent; color: black; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;"><a href="https://blog.singularitynet.io/shivom-partners-with-singularitynet-to-drive-genomic-medical-research-with-artificial-intelligence-bf93c804c879"> medical blockchain project.</a> But SingularityNET and Shivom aren’t going to democratize big medical data alone. What is needed is a wholesale redirection of medical data access, storage and analytics away from private collaborations between governments and megacorporations, and toward decentralized networks that are controlled by the world’s individuals in a democratic and participatory way. </span></span></div>
<div style="text-align: left;">
<b style="font-weight: normal;"><span style="font-family: "arial" , "helvetica" , sans-serif;"><br /></span></b></div>
<div style="line-height: 1.38; margin-bottom: 0pt; margin-top: 0pt; text-align: left;">
<span style="background-color: transparent; color: black; font-style: normal; font-variant: normal; font-weight: 400; text-decoration: none; vertical-align: baseline; white-space: pre-wrap;"><span style="font-family: "arial" , "helvetica" , sans-serif;">What is at stake here is, quite precisely: Who owns the deep knowledge about human life and death and health and disease and body function, which AI is going to discover by analyzing the world’s biomedical data over the next years.</span></span></div>
<br /></div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com5tag:blogger.com,1999:blog-11168555.post-46159585642555484952018-03-12T12:44:00.001-04:002018-03-12T12:44:36.764-04:00Machine Learning for Plant Disease Diagnosis and Prediction<div dir="ltr" style="text-align: left;" trbidi="on">
<span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;">AI is being applied to everything these days -- including various fields of endeavor generally thought of as low-tech and backwards, such as agriculture.</span><br />
<span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;"><br /></span>
<span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;">In this vein, I gave a talk last week in Leshan, in Szechuan province (mainland China), on the application of AI to diagnosing crop diseases (from images of leaves) and predicting disease course, disease response to treatment, etc.
In the talk, I reviewed a bit of existing literature and suggested some new twists based on discussions with farmers, crop doctors and agricultural researchers in Leshan.
</span><span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;">This was part of a collaboration between Chinese knowledge management firm KComber (and in particular their <a href="http://www.kcomber.com/sites/home/ccd..." target="_blank">Yoonop</a> service, </span><span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;">) and my bio-AI project <a href="http://mozi.ai/" target="_blank">Mozi AI Health</a> and decentralized AI project <a href="http://singularitynet.io/">SingularityNET.</a>
The talk I gave in Leshan wasn’t video-recorded, so after I got back home to Hong Kong I recorded a post-talk video going through the same concepts from the talk, using the same slides, with a few additions… Here it is!</span><span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;">
</span><br />
<span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;"><br /></span>
<div class="separator" style="clear: both; text-align: center;">
<iframe width="320" height="266" class="YOUTUBE-iframe-video" data-thumbnail-src="https://i.ytimg.com/vi/muiIk48TR_I/0.jpg" src="https://www.youtube.com/embed/muiIk48TR_I?feature=player_embedded" frameborder="0" allowfullscreen></iframe></div>
<span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;">
<i>(I was kind of half-asleep when recording the video as it was well past midnight but, that's when I found a free 30 min for this ... so it goes ... ) </i>
The slides from the talk, saved to PDF, are at: </span><a class="yt-simple-endpoint style-scope yt-formatted-string" href="https://www.youtube.com/redirect?q=http%3A%2F%2Fgoertzel.org%2FChengdu.pdf&redir_token=oPAiCE315afWIX4FKplUmfz_I4V8MTUyMDk1ODk4M0AxNTIwODcyNTgz&event=video_description&v=muiIk48TR_I" style="background-color: white; cursor: pointer; display: inline-block; font-family: Roboto, Arial, sans-serif; font-size: 14px; line-height: var(--yt-formatted-string-endpoint_-_line-height); text-decoration-line: none; white-space: pre-wrap;">http://goertzel.org/Chengdu.pdf</a><span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;"> .... (This PDF version lacks the robot videos I showed in the talk in Leshan, but those videos are somewhat peripheral to the main topic anyway....)</span><br />
<span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;"><br /></span>
<span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;">In more futurist-evangelist talks I give, I often stress the importance of using AI for broad global benefit -- because if early-stage AGIs are actively engaged in helping people of all sorts in various practical ways, the odds are likely higher that as these AGIs get smarter and smarter, they will be richly imbued with positive human values and interested in keeping on helping people.
Down-to-earth practical work on stuff like machine learning for diagnosing and predicting crop disease, is how this high level concept of "AI for broad global benefit" gets realized....</span><br />
<span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;"><br /></span>
<span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;"><br /></span>
<span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;"><br /></span>
<span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;"><br /></span>
<span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;"><br /></span>
<span style="background-color: white; color: #111111; font-family: Roboto, Arial, sans-serif; font-size: 14px; white-space: pre-wrap;"><br /></span></div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com57tag:blogger.com,1999:blog-11168555.post-40668767120998844962017-06-11T12:10:00.000-04:002017-06-15T05:31:33.146-04:00Qigong in Shanghai<div dir="ltr" style="text-align: left;" trbidi="on">
I'm on the way home from a consciousness conference in Shanghai — CSTS (Consciousness, Science, Technology and Society) conference. One of the more intriguing aspects of the conference, for me, was the significant number of Chinese researchers and practitioners there who were involved with parapsychology, qigong and other related areas…<br />
<br />
I’ve been reading for a while about a multi-decade project in Kunming dealing with macro-PK (psychokinesis), remote viewing and other capabilities among schoolchildren. I had emailed with one of the researchers involved with the project before, but hadn’t dug that deep. At CSTS I met a number of young Chinese researchers who were involved with gathering and analyzing data from the psychic children of Kunming; they showed me some videos of their work. These “kids” were quite matter-of-fact about their work — they had observed various children’s anomalous abilities many times, and viewed their role as documenting and studying the phenomena rather than trying to prove their reality (as they assumed the proof to have been completed well before their entry into the project). I’m planning to coordinate with these guys to visit Kunming sometime in the next year to check out these phenomena personally and see if I can help with experimental design, data analysis or theoretical analysis or whatever…<br />
<br />
On the other hand, one of the parapsychology-oriented presenters at the conference somewhat rubbed me the wrong way. He had a lot of exciting-sounding talk (in Chinese, translated for those of us in the audience who needed it) about channeling the mind of the universe, and so forth. He then offered a demonstration of his system for teaching people to channel the mind of the universe… and brought up some of his young female students to the front of the room. One of them proceeded to recite a short poem in an ancient Chinese style. He then noted the complexity and difficulty of writing a poem in this archaic formal style, and said that no modern youth could do that on their own — it could only be done by channeling the minds of the ancients, via the mind of the universe. Later on, this same teacher demonstrated qigong healing on an elderly man with serious stiffness in the fingers, due to a previously medically diagnosed problem. There was much talk about the cosmic wisdom of the universe and such, but the elderly man professed that at the end of the process, he felt a bit better but his fingers were still stiff. <br />
<br />
I did have the distinct sensation of seeing some sort of energy jumping up and down out of the elderly man’s body, at one point. On the other hand, it’s possible I was just sleepy and drifting off into some sort of dream.<br />
<br />
I’m not going to assert that this teacher was lacking in special abilities or engaging in fraudulent activities, or anything like that. <span style="background-color: white; color: #222222; font-family: "arial" , sans-serif; font-size: x-small;">But for sure, his demonstrations sent my Skeptic Sense way tingly.</span> According to the Chinese I asked, the student’s poem was competent but not especially exceptional. The amount of glowing evangelism about the channeling of the mind of the universe seemed far out of proportion to the magnitude of the phenomena displayed…. It didn’t escape me that the teacher was running a commercial school and seeking tuition-paying students…<br />
<br />
The next day the Kunming-project guys took me and Ruiting and another Hong Konger, L, to the Qigong Museum, a half-hour taxi ride across Shanghai. The historical displays there were fascinating, and the ancient illustrative artworks were ornate, complex and beautiful. <br />
<br />
Some random photos I took at the museum are here:<br />
<br />
<a href="https://photos.google.com/album/AF1QipMn2U9kiEkIVnCLZLI3Z2cGHCIFOuoAr8FKnGyx">https://photos.google.com/album/AF1QipMn2U9kiEkIVnCLZLI3Z2cGHCIFOuoAr8FKnGyx</a><br />
<br />
(My pictures come nowhere remotely near doing the museum justice; I didn't photo the stunning ancient drawing and such, but merely a few of the tech-y displays and machines.)<br />
<br />
There were also some intriguing electronic devices on display — electrical Qigong machines from the 1990s. These had a bunch of dials and knobs on them and seem to have been used to generate electromagnetic fields configured to affect the body in certain ways. The tour guide was quite negative about these devices, opining that qigong would always work better when manifested by a biological human being. <br />
<br />
And there were some photos of experiments done capturing qigong energy in various glass-bulb-type containers. I asked the tour guide what results had been obtained from this work, and he said there had been 10,000 papers published on this and he wasn’t qualified to summarize them, but there had been a lot of positive results.<br />
<br />
Ah, and some photos and information about the history of qigong-based anesthesia in Chinese operating rooms. It appears to have worked as well as chemical anesthetics, with fewer side-effects. Note that we don’t yet understand, within Western medicine, how chemical anesthetics work; the anesthetics in current medical use were figured out via trial and error.<br />
<br />
After our tour through the museum, a Qigong master — who had been at the consciousness conference, but had opted not to give a presentation to the crowd — offered to give us a practical demonstration. <br />
<br />
He asked L to stand still and relax and close his eyes. He then projected energy at L through his hand — and we watched, bemused and impressed, as he pulled L’s body back and forth, from a distance of about 12 inches. L (someone I know fairly well) tends to be skeptical of qigong, psi and other such things, and clearly was not complicit in the demonstration. When the demonstration was done and L opened his eyes, L was quite surprised when we told him his body had been swaying back and forth in coordination with the qigong master’s hand.<br />
<br />
The master then tried to do the same trick with me, but it didn’t work. I was standing there in a state of fairly deep relaxation, doing yoga/meditation breathing and so forth. When it was done, the master said it had failed due to a blockage of qi in my right knee, which he said had an old injury in it, from at least 5 years before. <span style="background-color: white; color: #222222; font-family: "arial" , sans-serif; font-size: x-small;">I don't know if this was actually the reason -- I'm a difficult case for hypnosis too. However, it's true that my right knee is mildly shaky due to a skiing accident I had 16 years ago.</span><span style="background-color: white; color: #222222; font-family: "arial" , sans-serif; font-size: x-small;"> </span> He then offered to try to cure my knee — and indeed, when he laid his hands on my knee in his special qigong way, the knee started feeling extremely hot and felt like it was vibrating deep inside at a high rate (a quite different feeling from what happens if the knee is physically shaken back and forth or if someone rapidly vibrates their hands on the skin).<br />
<br />
Afterwards, outside the museum, the master looked carefully at L’s body and correctly diagnosed a couple minor medical issues of L’s. He looked at me and noted a problem with my lower back, which so far as I know does not exist (my lower back feels fine and has never given me any problems so far); I did have an upset stomach from my dinner the previous night, though, which he did not note. I asked him to take a look at a woman who was there with me, and whom I knew to be in the 7th week of pregnancy — not far along enough to be showing at all. He looked at her and after a bit of scrutiny said “The baby is healthy, and your health seems good, congratulations.”<br />
<br />
Of course, none of this was skeptic-proof, but it was nonetheless pretty impressive and compelling. The impression I got of his qigong sensing and diagnostic process, was that it was a mix of things: careful observation and inference and medical experience, plus some sort of bioelectromagnetic field wizardry (involving both perception and manipulation of bioelectromagnetic fields, in ways that Western science does not yet encompass), plus probably some sort of psi component as well. <br />
<br />
Teasing apart these various aspects of his abilities is interesting to me, because I have a Western science oriented perspective. It’s interesting to me to figure out what aspects of what he does are due to psi, what aspects are due to weird bioelectromagnetic field dynamics, and what aspects are just due to him being a keen observer and a charismatic guy and a good doctor. <br />
<br />
On the other hand, from his point of view, it doesn’t really matter much which aspects of his practice fall into which of my Western analytical categories. Chinese philosophy and Chinese medical practice are holistic by nature, and what matters is the state of consciousness he achieves, and the medical cures he facilitates, rather than the explanation of what he does in terms of a particular combination of phenomena belonging to carefully-formalized reductive analytical categories.<br />
<div>
<br /></div>
</div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com99tag:blogger.com,1999:blog-11168555.post-84624812768755913262017-06-04T11:50:00.001-04:002017-06-04T12:05:30.529-04:00How To Save the World<div dir="ltr" style="text-align: left;" trbidi="on">
<div style="text-align: center;">
<b>How to Save the World: </b></div>
<div style="text-align: center;">
<i>The Biggest Risks Humanity Faces </i></div>
<div style="text-align: center;">
<i>and How to Militate Against Them </i></div>
<div style="text-align: center;">
<i>by Building a New Democratically-Oriented </i></div>
<div style="text-align: center;">
<i>Global Techno-Social Fabric</i></div>
<br />
<br />
<i>(Last week I had the opportunity to spend a few days brainstorming with my friends at the Economic Space Agency, an unusual and fascinating California startup organization. Among our technical conversations we also touched here and there on some of the major social issues facing humanity as it moves forward into the era of radically advanced technologies. This article summarizes some of the ideas we tossed around.)</i><br />
<br />
<br />
I generally tend to be an optimist, including about advanced technology and about the future potential for human and transhumanist growth. I think we’re going to create superhuman thinking machines and that some of us will merge with them and explore incredible new forms of mind, society and embodiment and experience. I think we will create radical material abundance that will end the era of working for a living, and end disease and death.<br />
<br />
I also think, though, there may be some serious downs as well as ups on the way to this radiant future. And there’s also a nontrivial risk that one of the downs takes us so far down as to prevent the amazing positive futures I envision from actually coming to pass.<br />
<br />
“Saving the world” as I mean it involves both the positive and negative aspects — making the world better by improving things dramatically, but also, preventing terrible things from happening and destroying the beautiful things we already have.<br />
<br />
On the positive side, there is a great diversity of human values around the world, but there is also some commonality. Nearly all of us have an innate sense of joy, and want to feel joyful. Nearly all of us have some sense of compassion, and want others around us to feel joyful as well. Nearly all of us want to be able to choose key aspects of our lives, and/or to have our families and communities able to choose key elements of their own paths. And many of us would like to grow beyond our current limitations, becoming more successful and more helpful and exploring new horizons. And fortunately, it appears likely that a variety of advanced technologies, currently already emerging, are going to be able to promote these positive values to an unprecedented degree. Computer networking, AI, biotech, nanotech, blockchain…<br />
<br />
And on the negative side, the list of risks we face is also well known. There’s the risk of natural disaster — comets hitting the Earth and so forth. There’s the risk of human-encouraged natural catastrophe — global warming causing an unforeseen chemical reaction in the oceans leading to massive release of poison gas, or some such. There’s the risk Nick Bostrom worries about in his book “Superintelligence” — that superhuman AIs will decide the molecules comprising humans can be more aesthetically or effectively used for some utterly nonhuman purpose. There’s the risk of World War III between nation-states. <br />
<br />
And then there’s the risk I think is actually the most worrisome — the risk that disaffected people, shut out of the centers of the world economy via radical economic inequality and/or political restrictions, leverage advanced technology to cause massive destruction. Which then creates a negative and chaotic global political situation, in which all sorts of destructive technologies get born (including maybe the ones Bostrom worries about).<br />
<br />
One thing I’m going to explain here is why I think this risk is a serious one — and then, what sort of things I think can be done to mitigate the risk. And the solutions I’m going to suggest are things that we really should do anyway, for a whole host of reasons.<br />
<br />
My logic here will involve a few decent-sized leaps, but I don’t think they’re <i>insanely</i> huge ones. You'll have to judge for yourself ;-)<br />
<br />
<b>When Robots Take The Jobs, Who Will Give Basic Income to the Residents of the Congo?</b><br />
<br />
Let’s suppose that AI and robotics and associated technologies keep on advancing impressively, so that the need for human labor in the economy keeps on decreasing. Then what happens to the people whose efforts are no longer needed in the labor force? In the developed world, we already see a strong movement toward universal basic income, which is pretty much the only rational and compassionate solution to the situation. But what about the developing world? Who will give universal basic income to the citizens of, say, the Congo or Central African Republic, when the developed-world economy has advanced sufficiently that outsourcing to the citizens of these nations doesn’t make any economic sense?<br />
<br />
One possibility is that developed-world philanthropists or governments rise to the occasion and distribute universal basic income throughout the world. But I don’t have much faith this will happen. <br />
<br />
One issue is that even the wealthiest philanthropists don’t have the funds to make a huge difference. Suppose Bill Gates gave half his wealth to supply Africa with universal basic income. 30 billion dollars would give 30 dollars to each African for one year. This is just not enough. No small set of good-hearted individuals is wealthy enough to make a real dent here. The whole class of super-wealthy individuals is rich enough to make a dent, but few of them care as much as the celebrated handful of billionaire philanthropists like Gates, Zuckerberg and Buffett.<br />
<br />
Another issue is that, in developed democracies like the US, average voters are typically very unhappy with foreign aid. Average Americans vastly overestimate the amount of foreign aid that the US government currently gives out, and they tend not to like it. When the US gov’t first starts giving out universal basic income, it’s not going to be enough to keep everyone happy. Average voters are not so likely to want to diminish their monthly payment in order to help faraway people with strange cultures and belief systems.<br />
<br />
It could happen that, due to advancing technology, supplying basic needs becomes SO cheap that it becomes politically unproblematic to supply a universal basic income globally. This might be the case if, for instance, we had a sudden breakthrough in molecular nanotechnology leading to fairly general-purpose, low-cost molecular assemblers. This would be awesome. But to be frank, I’m not counting on it. I’m sure these technologies will come, but they may come only a few years or a decade or two AFTER the technology that obsoletes outsourcing and leaves developing-world citizens economically stranded.<br />
<br />
<b>The (Extreme and Nasty) Qualitative Nature of Global Inequality Today</b><br />
<br />
For folks living in developed nations, it’s hard to get a sense of the extent to which people in the worse-off portions of the developing world feel (and are) left out of the modern world economy. (I’ve gotten a little bit of insight into this via frequent visits to Ethiopia, where I’ve been co-running an AI and robotics outsourcing shop since 2013. But I’m sure I still don’t have a full feel for these things, to the extent that I would if I’d grown up there.)<br />
<br />
Let me try to give a vague flavor of the situation. If you live in sub-Saharan Africa, the odds are that even if you can afford a smartphone, you can’t afford a lot of data minutes. Unrestricted viewing of, say, educational videos is not an option. <br />
<br />
You aren’t all that likely to be able to afford university tuition (if you do well on the exams at the end of high school, you have a good chance; but if you miss that one opportunity, your chance is probably nixed for life). <br />
<br />
If you want to start a business, good luck saving a couple thousand US dollars in seed funding. As soon as you save a few hundred dollars, odds are fairly high that some relative in a remote village will have a genuine urgent need for funding to pay for critical medical care. Your choice may be to spend your savings to help your great-aunt, or else to be responsible for her death. <br />
<br />
Do you want to use your hard-won savings to fly to some foreign tech hub to show your prototype to some investors? Good luck getting a visa. Sometimes it’s possible but it’s extremely chancy. (I have had many failures trying to bring my Ethiopian AI researcher colleagues to Hong Kong, Canada and the US on temporary visitor visas or work visas. These are people with advanced degrees in science and technology, and work history in AI software development. Occasionally it succeeds; often the immigration departments reject the applications with no reason given.)<br />
<br />
If you've managed to buck the odds and scramble like hell to gain the needed skills, maybe you can work online for overseas customers, doing consulting and saving money for your future that way. Well, except the e-work websites may well not be set up to allow you to register any payment method available to you on their sites. If you do manage to register, you'll get paid 1/10 the amount of folks in the developed world for doing exactly the same work, because of the country of residence indicated on your profile on the site. And oh, then you may find that your country's government decides to shut down the Internet for a week or two here or there (as I write this, Ethiopia has just shut down its whole country's internet for a week to try to avoid high school students cheating on their final exams) -- they don't seem to care what this will make your international customers think of your reliability! <br />
<br />
Some folks in the developing world fight through all these factors to achieve great or modest success. Many do not, despite having plenty of smarts and despite putting in MUCH more effort than the average, reasonably successful US, Western European or Chinese citizen.<br />
<br />
The situation in, say, the former Soviet republics or the less well-off countries in Southeast Asia or South America is not as extreme as in sub-Saharan Africa. But for many people it is qualitatively the same. Why do you think there are so many Filipino women willing to leave their spouses and children to work as domestic helpers in Hong Kong for US$600/month, half of which they’ll send home to their families? Why do you think there are so many attractive, college-educated young Eastern European women, willing to sell themselves as mail-order brides to unattractive, dull middle-aged men in the US or Western Europe?<br />
<br />
What’s at issue here is not economic equality per se — almost nobody I know is upset that some people are richer than others. There are differentials in ability and willingness to work that naturally lead to differences in wealth and income, at the current state of technological development. And almost nobody wants to wipe out all the impact of history on wealth — nearly everyone wants to be able to pass along some of what they create or earn during their lives to their children, for example. What’s at issue here is the ability to participate in a reasonably full way in the world economy. In other words, those developing-world citizens I know who are discontented with their role in the world economy (i.e. nearly all of them) are not discontented with the fact that there is an economic game with a competitive aspect to it, nor are they disgruntled with sour grapes that they have lost the game. They are discontented that, by dint of the nations they happen to have been born in, they are essentially disqualified from playing the game. They have to fight a hard battle with low odds, often entailing great personal and family sacrifices, just to get on the playing field of the international tech economy.<br />
<br />
I’m aware that life can be tough all around. Even the super-wealthy can find life a struggle each day, as newspaper tabloids amply document. I was raised middle class, nowhere near wealthy, and have worked quite hard all my adult life, with numerous ups and downs both personally and career-wise, some quite traumatic. I’m pretty contented now and I’m grateful for both the situation I was born into and various opportunities I’ve made and happened into. But I’ve seen close friends, also middle-class living in the developed world, commit suicide due to the difficulty of fighting through unfairly stacked competition and stultifying bureaucracy to realize their dreams. Human psyche and human culture involve suffering everywhere on the planet; and in some ways I feel like folks in the developing world have more satisfying lives than their materially wealthy counterparts, due to the rich and warm social fabric they so often weave. Complexities abound in nearly all human situations. But none of these complexities takes away the prevalence, unfairness or danger of the gross inequity in the current world situation.<br />
<br />
<b>Minimizing the Odds of Massive Inequality Leading to Global Catastrophe</b><br />
<br />
OK, so radical inequality of opportunity is prevalent and it sucks. Now take the next step, and realize that: In spite of these factors, there are more and more highly educated young people in the developing world, with understanding of advanced technology. Even the poorest nations are now equipped with computers and networks, and with biological lab equipment, and with universities teaching advanced science and engineering.<br />
<br />
It’s not hard to see what sorts of risks this situation leads to. If global inequality keeps increasing, and we have an increasing population of people who are largely shut out of the excitingly advancing global tech economy — but with significant access to modern education and technological tools — what do you think is likely to happen?<br />
<br />
As technology advances, it takes fewer and fewer people, with less and less intelligence and know-how, to create more and more destruction. This is true with computer hacking, it’s true with drones and robots, it’s true with biotech, and it will soon be true with nanotech.<br />
<br />
What we need to do, to prevent global wealth inequality and advanced technology from adding up to produce global catastrophe, is increase equitability of opportunity. We need to enable everyone in the world to have the opportunity to really play the modern global economic-social game. We need to all be in this together, on a basic level, or else, we’re likely to squander our chance to create radical abundance for all via destroying ourselves in a maelstrom of foolishness, selfishness and violence.<br />
<br />
I don’t mean to give the impression that the ONLY reason radical global inequality of opportunity is bad, is its strong potential to lead to widespread destruction. According to my own morals, this sort of radical inequality of opportunity is intrinsically a rotten thing, regardless of the risk it poses to global safety. I favor joy, growth and choice for all sentient beings, inasmuch is possible, and radical global inequality obvious is crappy on all three counts (it badly hurts joy, growth and choice, calculated in total across the globe). I’m just highlighting, in this particular essay, one among the many nasty implications of this sort of inequality: its reasonably high likelihood of fostering a situation in which vast numbers of humans get killed and nasty futuristic technologies get developed in the chaotic aftermath.<br />
<br />
Technology is not the whole solution here; this is a human psychology and culture problem as well as a technology problem. But I do believe that appropriate deployment of appropriate technologies can help create a context in which culture is more likely to evolve in a way that mitigates these problems.<br />
<br />
What technologies do we need? We need technologies that encourage a democratic global social fabric. That make it easier for people all around the world to connect with each other, to create media of all sorts for sharing with each other, to transact economically and emotionally with each other. That make it less and less practical for governments and large corporations — with their tendency toward impersonality and inertia — to clog, prohibit or pervert exchanges between individuals and the formation of ad hoc or persistent social networks of various sizes and types.<br />
<br />
<b>Six Critical Technologies for Enabling a Positive Future</b><br />
<br />
Fortunately these objectives can viably be achieved by a menu of technologies already in development to various extents. <br />
<br />
The following would be a good start:<br />
<br />
<br />
<ol style="text-align: left;">
<li>Mesh networks, so that Internet access is outside the control of large corporations and governments, and in more direct control of the people</li>
<li>Decentralized production of low-cost hardware. When we can 3D print the smartphones that enable the mesh network, in relatively low-cost local factories, then we’ll be in a pretty exciting position.</li>
<li>Machine translation, for both text and voice, that handles all the world’s languages. This is an area where my own research on artificial general intelligence may have a transformative role to play.</li>
<li>A social-network infrastructure that is widely used, that leverages machine translation and mesh networks, and that is out of the control of governments and corporations.</li>
<li>A decentralized, peer-to-peer economic exchange mechanism that is widely used and understood, operating on the mesh and requiring only prevalent low-cost hardware. Blockchain technologies provide an obvious basis here, including the new tech and ideas my friends at the Economic Space Agency are working on.</li>
<li>Videos, games, augmented reality, virtual reality, biofeedback, neurofeedback and AI interaction systems that are oriented toward helping people understand themselves and each other better. </li>
</ol>
<br />
<br />
Regarding the last point, another way to say it is: We need meaningful media that will help people grow, and help them come together, and help them confront the world’s difficult problems with love rather than hate — and we need this media distributed in a democratic and decentralized way rather than via advertising or propaganda dominated media channels. If provided with the technology to create and disseminate meaningful media, the people of the world will, in my estimation, most probably do so. There will be lots of less-meaningful media alongside, to be sure. But if media is more strongly separated from the desire of governments to spread propaganda and the desire of corporations to maximize profit, then the more positive and growth-oriented sides of humanity will have more chance to self-organize into powerful new configurations.<br />
<br />
These mechanisms would not automatically, magically make universal basic income spread globally. But they would make it far more likely. They would create massively richer, more positive interaction and interoperation between the developed and developing world economies. They would lead to the flourishing of creative new economic and social networks, including many oriented toward environmental, social and spiritual benefit. Rather than relying on governments or corporations, or majority vote in corporate media dominated democracies, to spread the bounty from technological progress widely — these tools would foster self-organization of decentralized mechanisms enabling individuals and small groups to reach out to other individuals and small groups and engage them in positive and mutually beneficial interactions.<br />
<br />
<b>Think About It ...</b><br />
<br />
Do we need all of the above to avert catastrophe? Not necessarily. <br />
<br />
Would the development of all this necessarily avert catastrophe? Not necessarily.<br />
<br />
But the more of this we can get, the more of a global “level playing field” we’re going to have, and the more of a sense we’re going to have that we’re all in this together.<br />
<br />
Development and deployment of this set of technologies would constitute a revolution, but not the violent kind. We don’t need to overthrow the world’s governments (though some are bound to get overthrown in the next couple decades anyway) and we don’t need to eliminate all the big corporations. What we need to do is to build new networks that join people together directly, in parallel to governments and corporations, and ultimately subverting their power and influence.<br />
<br />
With the above set of technologies, we would have a medium within which all sorts of new social, economic and cultural network would form. These would not lead to complete economic or social equality, and would not eradicate all the problems of the world. But they would go a long way toward enabling everyone on the planet to participate fully in the ongoing techno-social revolution. A world dominated by such technologies would be one in which positive new tech would have a higher odds of getting developed - including biotech that heals rather than biotech that kills, and AI that loves us rather than AI that repurposes all our molecules.<br />
<br />
It’s complicated. But it’s not more complicated than building Google, Baidu, the Internet, self-driving cars, New York or Beijing, or pulling off the Human Genome Project. Humanity can solve very complicated problems, when it focuses even a quite modest subset of its attention. <br />
<br />
Think about where we, as a society, are focusing our technology development efforts. Google and Tesla are great companies, for example. But how ultimately important is more effective online ad placement, and better creation of luxury cars? Why is our economy and society organized so that so many of the best educated brilliant minds on the planet are focusing their attention on such things? Of course, Google’s work on ads is funding its work on life extension (Calico); and Tesla’s work on luxury cars is funding its work on better batteries, which has very broad application. In general — in spite of the developed world focusing the majority of its economy on the creation of things that exacerbate inequality and appeal to the more selfish, shallow human emotions — we are still getting amazing and important things done. <br />
<br />
But it’s not clear that getting the important stuff done as a side-effect of building frivolous things to enrich or amuse the most wealthy is going to be good enough. <br />
<br />
It may be that we really need some more direct focus on technologies with strong direct potential for global good.<br />
<br />
Think about it.<br />
<div>
<br /></div>
</div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com57tag:blogger.com,1999:blog-11168555.post-16625140068731502472016-12-10T23:31:00.002-05:002016-12-10T23:45:22.799-05:00The Tech-Startup Attractor: Musings on the Thiel Fellowships, Singularity University, and the value of good old-fashioned universities<div dir="ltr" style="text-align: left;" trbidi="on">
<br />
This Business Insider <a href="http://www.businessinsider.com/peter-thiel-tried-to-prove-that-brilliant-kids-dont-need-college-2016-12" target="_blank">article on the outcome so far of the Thiel Fellowship experimen</a>t is interesting, though not surprising..<br />
<br />
As you may recall, <a href="https://www.entrepreneur.com/article/234544" target="_blank">Peter Thiel launched the Thiel Fellowship program as a "20 under 20" initiative,</a> with a stated aim of showing that - for bright ambitious youth anyway -- college is not necessary and is in many ways not the best way to spend 4 years of one's young adulthood. The Thiel Fellows were each given $100K over 2 years, with a goal of supporting them in pursuing their own thoughts, dreams and visions....<br />
<br />
As the Business Insider article reports, as the Fellowship experiment has continued for a few years, it has evolved a bit ... in the beginning it seemed like it was going to focus broadly on ambitious and brilliant youth with all sorts of creative new ideas and direction, and on giving them space to flesh out their thinking without needing to worry about paying the bills ... but it seems to have gravitated more toward a sort of "social network for young entrepreneurs", focusing mostly on young people with tech business projects reasonably likely to create near-term profit ... including many who are already having significant business success. And the main value-add of the program is coming out to be, not the cash stipend, but rather the social network to which the Fellowship gives access.<br />
<br />
All of which is great, and surely moves technology and business and society forwards a bit. However, it does not whatsoever show that dropping out of college is a great path forward for youth in general. What it shows is more like: IF you are young and want to start a tech biz based on an idea that appears to the Silicon Valley tech community to have significant near-term financial potential, THEN dropping out of school and into an extremely influential social network (well-connected with a host of high-net-worth individuals and impactful tech companies and VC funds etc. etc.) is a damn good idea, if you get the opportunity...<br />
<br />
Well, yeah.... But this doesn't really say much about the pluses or minuses of going to college or getting a college degree if you DON'T have the opportunity to get rapidly embraced by a world-class social network like this...<br />
<br />
I'm not especially an apologist for the contemporary university system, which annoys me in many ways, with (among other problems) its focus on rote learning, its obsession with dividing knowledge into irrelevant disciplinary bins, and its tendency to squelch individual and group creativity. <br />
<br />
On the other hand, I have to admit that universities are the one area in human society that has consistently, over a long period of time (nearly a millenium!), provided an environment in which learning new things and developing radical new ideas is generally encouraged, apart from the short-term reward that such learning and ideas may bring. All the egregious flaws of the university system aside, this is not to be scoffed at.<br />
<br />
Society gives all of us a lot of pressure to pursue short-term reward in various ways, and on various levels. Even as universities come to focus more on career-preparation majors, and professors are pushed to pull in grant funding rather than work on obscure or out-there topics that funding agencies ignore -- still, compared to the other aspects of our society, universities seem by far the MOST supportive of learning and creation and invention not tied to short-term reward.<br />
<br />
Of course there are non-university institutions that out-do universities in this regard, but they are small and scattered and end up not being accessible to most people.<br />
<br />
Of course, nearly anyone in the developed world can find other ways to spend their time learning and creating, without enrolling in university. But we are all susceptible to various social pressures, so -- even with all the information and communities available on the Internet -- it is still valuable to have a physical environment where learning and creation are core to the mission and vibe.<br />
<br />
One point I often end up making in conversations about AI is that every one of the "deep neural net" algorithms being used by big tech companies these days, was invented by university professors and published in the academic literature. Then the big tech companies took these (often via hiring said professors or their grad students) and implemented them more scalably and got amazing practical results. But the core deep learning algorithms were invented in the university setting not the tech company setting, and they were invented alongside thousands of other algorithms, none of which had widely obvious commercial value at the time they were invented. Many of these other algorithms will never prove practically valuable; some may ultimately prove far more valuable than deep neural networks.<br />
<br />
Quantum computing obviously is the same way. Where was quantum theory developed? And where were the original ideas underlying quantum computing worked out? Where are the speculative designs and lab experiments and math papers being done today, that are laying the groundwork for the quantum computers we'll have in our compute clouds 15-25 years from now ... for the Quantum Processing Units (QPUs) we'll have in our smartphones and smartwatches, in our robots' brains, maybe implanted in our own brains. <i><b>Hint: mostly not in venture-funded startups, nor in the labs of big tech companies... </b></i><br />
<br />
The Thiel Fellowships are a cool program, but they don't seem to be fostering the kind of wide-ranging intellectual exploration and concept creation that universities -- in their screwed-up, contorted and semi-archaic way -- have so often fostered. Rather they seem to be fostering some young people to do what Silicon Valley does best -- take ideas already formed by other folks and commercialize and market them, make them scalable and slick. I don't want to discount this sort of work; I love my Android phone and Macbook and Google Search and all that too.... But this is a very particular sort of pursuit, and the fact that getting embedded in an awesome social network is more useful than university for this sort of thing, is pretty bloody obvious, right?<br />
<br />
I see some parallels with how Singularity University (the non-degree granting educational organization, founded by Ray Kurzweil and Peter Diamandis and others in Silicon Valley) has developed. While I'm currently an advisor for SU's AI and robotics track, I'm not that intensively involved with the organization these days. However, I was fairly heavily involved with SU when it was founded, and before it was founded. <br />
<br />
The initial legwork for putting SU together was largely done by Amara Angelica (who runs KurzweilAI.net for Ray) and Bruce Klein, who at the time was working for me as President of Novamente LLC. I was paying Bruce a modest salary for his Novamente work, which covered his basic bills while he spent 6 months doing social networking trying to put together the founding meeting for Singularity University. The founding meeting -- which I ended up not attending as I was busy with so much other stuff -- was a big success ... Ray Kurzweil and Peter Diamandis shared their vision wonderfully and recruited the needed founding donations from the various individuals Bruce and Amara and their colleagues had gathered together (with help from Ray and Peter as well) ... and SU was off to the races ...<br />
<br />
When Bruce and Amara and I were first talking about SU, however, our discussions had a pretty strongly Singularitarian vibe. We were talking about how to radically accelerate progress toward AGI, mind uploading, Drexlerian nanotech, radical longevity, and so forth. And in the first couple years of its existence, SU was fairly much in this vein, though already a bit more "startup bootcamp" oriented than we had been thinking.<br />
<br />
Looking at SU now, it's awesome for what it is -- but it's become far more focused on short-term hi-tech business opportunities than it initially seemed would be the case. SU has done a heap of good for the world, by bringing future-minded entrepreneurs and others from all around the world together, to social network with Silicon Valley tech leaders and brainstorm on how to create new startups using advanced tech to improve the world. <br />
<br />
And much as with the Thiel Fellowship, I believe the main value-add SU has ended up providing to its students is the social network. Definitely, for a future-oriented business executive or scientist from Quatar or China or Bolivia or Ethiopia, the chance to get to know dozens to hundreds of Silicon Valley and international tech-biz geeks can be pretty priceless...<br />
<br />
Perhaps much of what we see in both the Thiel Fellowship and Singularity University cases is merely the power of the "tech startup attractor" for programs based in Silicon Valley. Silicon Valley is damn good at tech startups, and is not necessarily equally good at providing alternative means of giving young people broad education or space to wild-mindedly create new ideas ... nor at encouraging people to make huge leaps toward the Singularity in ways that don't promise short-term business success.... <br />
<br />
Of course, Silicon Valley doesn't have to be everything -- it's a big world out there, with lots of wealth and brilliance and capability in so many different places -- and it's incredibly impressive what things like the Thiel Fellowship and Singularity University are contributing to the world. But the relatively small role these sorts of things play in the bigger picture of what's going on on the planet, should also not be lost sight of...<br />
<br />
So far, universities are still pretty damn useful, in terms of providing environments for young people to learn how to learn, and space for young people to create and grow without the world's usual pressures.... <br />
<br />
And so far, the challenge of directing significant resources to really ambitious Singularitarian goals like AGI, mind uploading and Drexlerian nanotech, has yet to be met.... We are moving toward these goals anyway, and progress is excitingly fast by historical standards. Yet it seems to me that we could be progressing faster and in many ways more beneficially and interestingly, if not for the tendency of more visionary initiatives to get sucked into attractors related to short-term profit-seeking.... <br />
<br />
And it's also clear to me, that, in our current path toward a radically better future, good old traditional universities are continuing to play a very central role, in spite of all their archaic peculiarities. I would love for far better modes of social organization to emerge, but this process is still underway; and currently the Silicon Valley tech-startup network -- with all its diverse and fascinating manifestations -- is more a complement to traditional universities than an alternative...<br />
<br />
<br />
<br />
<br />
<br />
<br /></div>
Ben Goertzelhttp://www.blogger.com/profile/01289041122724284772noreply@blogger.com27