Tauchain through the Moravec Prism
blockchain·@karov·
0.000 HBDTauchain through the Moravec Prism
<html> <p>Hans Moravec [1] is the patriarch of robotics [2]. The real one, not the Sci-Fi father. Asimov was just the prophet in this scheme of things. </p> <p>Moravec to Kurzweil is what's Bitcoin to Ethereum and Satoshi to Vitalik. </p> <p>Sorry, for the rough joke. No offence, Ray! Back in the earler 2000s I bought your books too [3]. </p> <p>In my humble opinion - aside from the ''reality intratextualization'' [4] concept - the other wisdom jewel of Moravec's - fruit of a life devoted to robotics - is the Moravec's Paradox [5]. </p> <p>Explained in his own words: </p> <blockquote><code>Encoded in the large, highly evolved sensory and motor portions of the human brain is a billion years of experience about the nature of the world and how to survive in it. The deliberate process we call reasoning is, I believe, the thinnest veneer of human thought, effective only because it is supported by this much older and much more powerful, though usually unconscious, sensorimotor knowledge. We are all prodigious olympians in perceptual and motor areas, so good that we make the difficult look easy. Abstract thought, though, is a new trick, perhaps less than 100 thousand years old. We have not yet mastered it. It is not all that intrinsically difficult; it just seems so when we do it.</code></blockquote> <p>or with Steven Pinker's [6]: </p> <blockquote><code>The main lesson of thirty-five years of AI research is that </code><code><strong>the hard problems are easy and the easy problems are hard</strong></code><code>. The mental abilities of a four-year-old that we take for granted – recognizing a face, lifting a pencil, walking across a room, </code><code><strong>answering a question</strong></code><code><em><strong> </strong></em></code><code>– in fact solve some of the hardest engineering problems ever conceived... </code></blockquote> <p>As I noted in a previous related post of mine [7], a system's value dynamics is all about how it scales. Preferable of course are systems which make more good to go around than less. Respectively, to come around. </p> <p>Humanity is a network, and its scaling is stumbled by our innate attentional resources limitations. </p> <p>Human social interaction is a skill and we naturally have only as much of it. </p> <p>For now, in the good old hierarchic way [8], we can't deny that we scale satisfactory well (as compared, lets say, to our DNA-blockchain-fork-out first cousins the chimps [9]) for collaborating efficiently on successful execution of trivial tasks like empire building or colonization of the Galaxy. </p> <p>But not all problems we encounter are simple. In fact most problems are more complex than we are capable to grok and master in the hierarchic collaboration mode, which quickly slams into the Shannon's 'brick wall' [10]</p> <p>Ohad Asor's Tau [11] is intended to be humanity upscaler [12]. This project is the first and only one I've discovered so far where the so obvious (after you know it) problem is even identified, stated and addressed. </p> <p>This means uplifting the individual humans too, because we are literally AIs serially manufactured by our society (cf. feral children [13]). </p> <p>It feels easy for us to attend, to remember, to forget, to think, to talk, to work together - so it is extremely Moravec-hard! </p> <p>Tau is unique approach towards the Moravec-hardness of these problems in the realization that we do not need at all to waste time and resources to mimic nature and copy ourselves and to create high tech homunculi [14]. </p> <p>The 'problem' is the solution. Don't 'solve' it - just god damn use it! </p> <p>It is the people who ask questions, upload statements, express tastes and do all that qualia [15] crap humans usually do. </p> <p>The machine distills the semantic essence of all the shared thought flow, treats it as wishes specs, and automatically converts into executable code, incl. its own code self-amendment. </p> <p>As Moravec found out few decades ago [16] : </p> <blockquote><code>The 1,500 cubic centimeter human brain is about 100,000 times as large as the retina, suggesting that matching overall human behavior will take about 100 million MIPS of computer power. </code></blockquote> <p>When these processing brain things are really put together <strong>in numbers</strong> the result is unprecedented power. An unstoppable force. A glimpse into it by Ohad [12]:</p> <blockquote><code>It turns out that under certain assumptions we can reach truly efficiently scaling discussions and information flow, where 10,000 people are actually 100 times more effective than 100 people, in terms of collaborative decision making and collaborative theory formation. But for this we'll need the aid of machines, and we'll also need to help them to help us.</code></blockquote> <p>Without application of dehumanizing individual upgrades, without to be necessary to understand and reengineer the billions of years of evolutionary capital, but just harness it and use it. (Scaling itself must be scalable, too, ah?) </p> <p>In my personal up to date limited understanding it seems that it is indeed the HUMANITY what's to be known as the Tau's 'Zennet Supercomputer', and the machines are the ... collab amplifier media, the 'internet' of it. (Ohad, correct me if I'm wrong, please.) </p> <p>Like laser configurations of minds. </p> <p>With performance stronger than thought. </p> <p><em>NOTE: I have the honor to be in the Tau Team, but all reflections in this post are personally my opinion.</em> </p> <p><img src="https://www.jetpress.org/volume1/All_things_075.jpg" width="614" height="536"/></p> <h3>References: </h3> <ul> <li>1. https://en.wikipedia.org/wiki/Hans_Moravec</li> <li>2. https://www.ri.cmu.edu/ri-people/hans-moravec/</li> <li>3. http://singularity.com/</li> <li>4. http://www.orionsarm.com/eg-article/47645c107aed6</li> <li>5. https://en.wikipedia.org/wiki/Moravec%27s_paradox</li> <li>6. https://en.wikipedia.org/wiki/The_Language_Instinct</li> <li>7. https://steemit.com/blockchain/@karov/geodesic-by-tau</li> <li>8. https://en.wikipedia.org/wiki/Network_architecture</li> <li>9. https://phys.org/news/2011-10-humans-tasks-chimps-dont.html</li> <li>10. https://en.wikipedia.org/wiki/Claude_Shannon#Information_theory</li> <li>11. http://www.idni.org/blog/agoras-to-tml</li> <li>12. http://www.idni.org/blog/the-new-tau</li> <li>13. https://en.wikipedia.org/wiki/Feral_child</li> <li>14. https://en.wikipedia.org/wiki/Homunculus</li> <li>15. https://en.wikipedia.org/wiki/Qualia</li> <li>16. https://www.jetpress.org/volume1/moravec.htm</li> </ul> <p><strong>18qSKUUTAGw1uL53simrSiZ6pJpfxKACvj for research support. Thanks. </strong></p> <p><strong>Copyright © 2018 Georgi Karov. All rights reserved.</strong></p> </html>