Re: [agi] Re: Information MetaCriterion

Noise and error. That's where all the answers are, if the agent can afford to redirect and expend computational resource, if the metacriterion convinces though a perception of potential gain... muscling past the chaotic inflection leveraging directed self-organizing potential. --

Re: [agi] Re: Information MetaCriterion

is this HOMO x thing a tasteless joke about everyone being gay? -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T0fc0d7591fcf61c5-Mfac5755a5492c1ae62981173 Delivery options: https://agi.topicbox.com/groups/agi/subs

Re: [agi] Re: Information MetaCriterion

Now Matt's rolling. Yeah.  -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T0fc0d7591fcf61c5-M445a9421e6e39904cfadbd9e Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Information MetaCriterion

On Fri, Nov 22, 2019, 10:14 PM Alan Grimes via AGI wrote: > =\ > > I really see turning the earth into goo as strongly undesirable. =| It depends. The extinction of Homo Erectus and evolution into Homo Sapiens is generally seen as desirable, at least to the surviving species. Once we can engin

Re: [agi] Re: Information MetaCriterion

its pretty much goo'd already,  lots of sick cunts everywhere getting off on travestic things,  i think we need another noahs flood. -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T0fc0d7591fcf61c5-M6479bf9cf0f0e

Re: [agi] Re: Information MetaCriterion

immortal.discover...@gmail.com wrote: > Don't like change? :) Everything dies. Your body/room is changing at > the particle level as you read this. Physics is taking its course. > Things are predictable. Nanobots are coming I'll make sure of it > 100%! Your land and valuables are just your idea

Re: [agi] Re: Information MetaCriterion

Sight-seeing feels good because you're eating data. Games let you try things that are unsafe, impossible, etc. You can be others. Use any tool, increase their size, speed up... -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.co

Re: [agi] Re: Information MetaCriterion

Don't like change? :) Everything dies. Your body/room is changing at the particle level as you read this. Physics is taking its course. Things are predictable. Nanobots are coming I'll make sure of it 100%! Your land and valuables are just your idea of staying alive - your food, wives, homes

Re: [agi] Re: Information MetaCriterion

=\ I really see turning the earth into goo as strongly undesirable. =| I mean find something that doesn't currently have any realestate value, goo that! Nobody will care. But places like: https://wvtourism.com/town-guide-elkins/ are seriously fucking scarce in this universe, so should be valued a

Re: [agi] Re: Information MetaCriterion

On Friday, November 22, 2019, at 3:53 PM, rouncer81 wrote: > I.D..   what takes more to store >  A perceptron take key data/ Like the book ends in a piece wise function. Than synthetic data can  be generated when needed for what ever is in between. for lossy compression. But more book  end in o

Re: [agi] Re: Information MetaCriterion

I know a file can be compressed. I'm talking about de-compression in the universe. Larger systems that have larger context are more likely to survive. A utopian sphere can fight a small pirate UFO. Small systems are more likely to have less junk and are therefore the first systems that are the m

Re: [agi] Re: Information MetaCriterion

I.D..   what takes more to store,  every single picture frogger makes, or the frogger code that generates the pictures? -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T0fc0d7591fcf61c5-M01735ca1282484865474df3e D

Re: [agi] Re: Information MetaCriterion

Bigger systems are more powerful than small systems. Smaller is better, bigger has more junk, but once there's no junk it will crush the small system. -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T0fc0d7591fcf

Re: [agi] Re: Information MetaCriterion

 I am talking to immortal and the others. I completely reject you as a t bit chomp. -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T0fc0d7591fcf61c5-M471d086b7c797daab9654e76 Delivery options: https://agi.topicbo

Re: [agi] Re: Information MetaCriterion

Schmidhuber's Speed Prior-Based Inductive Inference may be the most principled way of introducing time but it seems to lack a principled way of making bits and time commensurable so he just punts and says "Kt" which has the dimension of informati

Re: [agi] Re: Information MetaCriterion

 Two inductions. First is a pattern loop. The second is a fragment of a pattern loop called the "launch".  Launch is transition into another pattern loop or launch.  The pattern of the universe from the end is one big launch. Or when a AGI become conscious to the  time it turns off. Or even the l

Re: [agi] Re: Information MetaCriterion

Thermophiles! https://www.nps.gov/yell/learn/nature/life-in-extreme-heat.htm https://blog.nationalgeographic.org/2013/08/02/5-extreme-life-forms-that-live-on-the-edge/ -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/a

Re: [agi] Re: Information MetaCriterion

Ok so water bears cheat a bit, they become hard by vitrifying. But that isn't the only or best trick to use: https://en.wikipedia.org/wiki/Extremophile -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T0fc0d7591fcf

Re: [agi] Re: Information MetaCriterion

So why are all our thingies so small? Why aren't birds as big as a tree? Why aren't men as big as a building? Why aren't programs a billion lines of code long? Why aren't cells as big as an apple? Why aren't we inviting other workers into our room? Too much clutter? They get in the way, there's

Re: [agi] Re: Information MetaCriterion

I believe the deeper question is why Occam's Razor works so well across every branch of science. What makes Solomonoff induction the core principle of machine learning? Why is it so successful at prediction in every domain? Why are theories that require fewer words or symbols to describe more likel

[agi] Re: Information MetaCriterion

I will agree however that if there is a principled way to include both time and space as constraints in a model selection criterion it makes pragmatic sense at the very least because what one is trying to do is predict which by definition is in time. On Thursday, November 21, 2019, James Bowery w

[agi] Re: Information MetaCriterion

In the context of model selection one is talking about the smallest among various executable archives that exist. On Thursday, November 21, 2019, TimTyler wrote: > On 2019-11-21 11:46:AM, James Bowery wrote: > >> I, quite deliberately, did not mention "Solomonoff Induction" as an >> information

Re: [agi] Re: Information MetaCriterion

Are you saying, a monkey does the same job a human does - ride a bicycle, but the monkey is smaller in size and can swing from branches as well, so we choose it instead? We are picking and choosing something based on reasonshow lossless it is, how small it is, how long it takes to run, how

Re: [agi] Re: Information MetaCriterion

On 2019-11-21 11:46:AM, James Bowery wrote: I, quite deliberately, did not mention "Solomonoff Induction" as an information criterion for model selection, precisely because it is not computable.   The point of my conjecture is that there is a very good reason to select "the smallest executable

Re: [agi] Re: Information MetaCriterion

If I can spawn a finite but unlimited number of parallel processes in "space", I can compute AIXItl, for example. So let's say the generating space is projected down into 3D space + time -- it is approximated by time, correct? In other words, once you admit "space" as a computation dimension, don

Re: [agi] Re: Information MetaCriterion

On Thu, Nov 21, 2019 at 3:33 PM Matt Mahoney wrote: > That's a good example of the weakness of the lossy compression model. We > only use it because it's better than nothing. > Ah, but my conjecture is that selecting the executable archive _is_ better than _any_ lossy compression model.Moreo

Re: [agi] Re: Information MetaCriterion

On 2019-11-21 11:46:AM, James Bowery wrote: The point of my conjecture is that there is a very good reason to select "the smallest executable archive of the data" as your information criterion over the other information criteria -- and it has to do with the weakness of "lossy compression" as mo

Re: [agi] Re: Information MetaCriterion

That's a good example of the weakness of the lossy compression model. We only use it because it's better than nothing. Find a function that approximates the data. Choose the function that minimizes the description length of the function plus the prediction errors. Another example. Suppose you have

[agi] Re: Information MetaCriterion

I, quite deliberately, did not mention "Solomonoff Induction" as an information criterion for model selection, precisely because it is not computable.   The point of my conjecture is that there is a very good reason to select "the smallest executable archive of the data" as your information crit

[agi] Re: Information MetaCriterion

Erratum:  if not without loss, then, at least, without loss **except* *whatever the particular criterion deems "noise". -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T0fc0d7591fcf61c5-M224ee444c845662e7228d1da D