top AGI leads 2018½

hi all, earlier this year released a new theory of AGI on this blog. it got substantial/ gratifying hits and am still pursuing it. aligned with this work, did a massive survey of existing state-of-the-art AGI leads. my initial idea was to try to summarize/ survey the different approaches. still have that in mind but its a almost herculean task and too much to bite off at this moment. this blog gets respectable hits but the audience is very spread out and not vocal/ participatory, dampening some of my energy for that high effort currently (but also not ruling it out).

however, this is a massive step in that direction, just painstakingly collecting this large ~180 link set/ collections of top leads.

much of this was found via the MIT AGI slack channel. its like trying to keep up with a firehose, but its very lively and cutting edge and also with tons of noise. as an expr goes used in this blog on various occasions, not for the fainthearted!

in compiling this its striking to me how both/ simultaneously brilliant and obscure some of these approaches are. some seem to me to be very much getting at the heart of AGI (and realized they are closely aligned with my own) but like my own audience, there is a lot of scattering. so far there is fairly low coalescing/ coalescence of groups around common themes/ consensus. my feeling is this disconnection may fall dramatically in the coming years esp with widely known/ publicized breakthrough(s) that drive the currently somewhat meandering herd down much more specific directions. it will be challenging-to-difficult but not inconceivable, exactly that happened on a substantial scale with deep learning within the last few years.

while it may seem overwhelming/ insurmountable at times, in some ways the AGI problem purely reduces to an architecture/ coordination problem, aka engineering. and notice some groups are arriving at the same answer from different directions (mainly psychology, (neuro)biology, machine learning, statistics/ data science/ big data, education/ learning theory, robotics, game AI, etc), with different languages/ vocabularies/ terminologies/ paradigms that are showing some/ early signs of converging/ convergence.

with new technologies, its all about “traction + momentum”. within the next few years, am expecting some major strategy/ consensus to emerge that builds on deep learning that gives rise to a plausible path/ route to AGI. have already outlined it myself, and think my ideas are close to the “secret sauce”, but my influence is low. fully expect nearly the exact same ideas to gain major traction but when espoused by some other leading light/ monolithic authority in the field, either an organization or individual or some combination of the two. it will likely be in the form of some step from the following ideas toward the more specific/ “laser-focused” direction.

odds are if there is some major AGI theory circulating at the present time, its pointed to in these refs, the well-known and not-so-well-known. and boldly both going out on a limb with a crystal ball, furthermore, think odds are strong that a “correct/ viable” AGI theory is in the not-too-distant future/ intermediate horizon and that the seeds will be contained (“holographic like”) in at least some refs cited here, maybe even many.

Continue reading

Advertisements

collatz shift

this months title is hopeful and in line/ theme with the last entry title. the disordered climbs were found to be a major wrench in the works of many months of analysis. however there are now some signs that even though seemingly without signal, maybe there are some angles to leverage on the disordered glides. am starting to get some general ideas. have developed/ honed some even stronger tools/ techniques. will they be enough to crack the problem?

this code is a modification of review51 to extract the recent/ latest generated glides from mix25g. it turned out extracting usable climbs from the table/ database was really nontrivial. 1st this code looks for the most common bit width encountered in the glides (climbs). this ends up to be exactly 200 the same as the bin count. this is interesting and maybe worth exploring further: all the lower glides are tending to cross (repeatedly) through that section. there are ~10K total iterations. then the climbs containing at least 1 iterate with that width are selected.

Continue reading

FLUID PARADIGM SHIFT 2018½

hi all. what if research into the interpretation of QM leads to a QM + GR path/ direction for unification? that is exactly the ambitious, overarching, but not inconceivable promise of the fluid paradigm of physics.

its been ~¾ a year since a last bold/ ambitious blog on copenhagen interpretation and new fluid ideas in quantum mechanics and physics.

at this point have blogged about ½ decade on some of these subjs and my neurons are really buzzing, crackling and snapping lately at full volume, the field is going thru an identifiable paradigm shift predicted years ago in this blog.

have collected a copious collection of new info/ leads, my links really runneth over. it really looks like nearly critical mass in some ways.

have been waiting to blog on this a few months and looking for an opportune moment. am expecting some massive signs to show up. many have already shown up. am finally deciding to write all this up at the near ½ year point.

one of the biggest signposts/ BREAKTHRUS is the new Becker book, What is Real, the Unfinished Quest for the Meaning of Quantum Physics.[a] havent bought it yet but its on the top of my to-read pile. this is causing big, maybe even massive waves in the mass/ popular media eg NYT but also the scientific journals such as Nature. its being reviewed by top experts positively. major response on social media such as reddit also. maybe a gamechanger.

Continue reading

collatz new highs

am still clawing/ scrounging for any “big picture” leverage, resuscitating old leads. after some extended musing of new findings, came up with this latest idea. there are lsb triangles even in the disordered climbs for the uncompressed mapping. do these mean anything? seemed to see some cases/ trend where the triangle sizes successively decrease in the climb, ie dont successively increase. can this be quantified? this is somewhat similar to the old workhorse “nonmonotone run length”. instead, its something like “count of new highs” (in bit run lengths either 0/1). that statistic is (maybe not uncoincidentally) used in stock market analysis, which has to deal with some of the most wild/ intractable fractals of all.

it was not hard to wire up the last “dual/ cross-optimizer” (ie both within fixed bit sizes and also increasing bit sizes) to calculate this metric, named here ‘mc’, serving its intended purpose of trying out new ideas quickly. ran it for 100 bit width seeds and then it did indeed seem to flatline somewhat. upped the seed size to 200 bits and then more of a (very gradual) trend is apparent. it looks like a logarithmic increase (‘mc’ red right side scale, other metrics left side scale). ‘mw’ is the # of iterations since last max/ peak run length. the optimizer ran for a long ~650K iterations.

Continue reading

collatz killer

hi all here we go with the latest installment. trying to come up with new names/ themes. again theres a “pivot” going on at the moment but maybe there are now too many too count. time now for an intermediate retrospective/ pov. my recent physics blog talked about “killing the copenhagen interpretation” and thats my latest idea for this problem. the problem is definitely “killer” in many senses of the word. it kills all great ideas launched against it, its like an impenetrable fortress.

there was a tone of optimism in a lot of prior writing. now looking all that over, it was based on a longtime theme that was yielding fruit(s) of labor(s). the basic idea is that there are locally computable “features” that can, with enough ingenuity, predict longterm glide behavior with high accuracy, and also generally explain other basic trajectory dynamics properties. this clearly ties in with the machine learning approaches. this research theme has been pursued for several years now.

however, last month there was a massive setback on this particular theme/ direction. did you catch it? to summarize, the features being used, mostly based on (binary) density, were leading to a lot of insights and leverage on the problem. but there was a moment a few years ago when the research started to focus on generating density-based seed trajectories instead of more generally. that turned out to be a major detour bordering on a mistake (in 2020 hindsight). 😮 😳 😥 😡 👿 o_O

Continue reading