Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Karl Friston: a neuroscientist who might hold the key to true AI

wired.com

Post reply on HN
Print view View original

101–110 of 111 posts

Re: Karl Friston: a neuroscientist who might hold the key to true AI

#101

Post by klodolph »

lurquer wrote: post #97
klodolph wrote: post #77

Earlier quoted context omitted.

This is a completely minor point here... but "fallacy" means that there's something wrong with the argument, if you have a disagreement about facts or assumptions then the word "fallacy" doesn't really apply (you can just say "wrong" instead).

It suffers from the fallacy of petitio principii, in that it assumes arguendo that consciousness is comprised of neurons (and, as mentioned above, that a neuron is less complex than consciousness.) But it's not stated as an 'argument' in any case so perhaps the term fallacy was out-of-place.

Petitio principii is when the premise assumes the truth of the conclusion, but since there is no argument and no conclusion it's impossible for the statement to suffer from that fallacy.

Re: Karl Friston: a neuroscientist who might hold the key to true AI

#102

Post by toxik »

Rainymood wrote: post #27
toxik wrote: post #13

Earlier quoted context omitted.

With all due respect, one sentence explaining how you think the mind works isn't really worth much. It doesn't amount to much more than "the brain tries to explain reality." Yes, ok, but how do you translate that into some algorithm? How does it relate to gradient descent methods on neural networks?

>With all due respect, one sentence explaining how you think the mind works isn't really worth much. With all due respect, one sentence can be worth a lot. Some examples: > F = MA Another > E = MC^2 And another > G_{\mu, \nu} = 8 \pi G (T_{P\mu, \nu} _ \rho_{\Lambda} g_{\mu, \mu}) Another example >To be, or not to be; that is the question; Et cetera, et cetera. The length of something does not necessarily imply that…

None of these mean anything without their respective context. In fact, they're all pretty pedestrian taken as a single sentence.

Re: Karl Friston: a neuroscientist who might hold the key to true AI

#103

Post by Chris_Jay »

sixdimensional wrote: post #96
ArtWomb wrote: post #78

Most grad-level Deep Learning classes have a week or so devoted to "Approximate Bayes" methods. And it's conceivable future updates to all popular probabilistic programming languages will include "programmable" rather than "fixed-function" inference methods. "Inference Metaprogramming" paper https://people.csail.mit.edu/rinard/paper/pldi18.pdf Latest state-of-the-art research will be presented at upcoming NeuroIPS co…

Since you mention Bayesian methods, I thought I may randomly ask you - have you come across any good work about applications of subjective Bayesian statistics in AI? I was particularly interested in subjective Bayes theory due to the way it seems to interleave human input with mathematical theory. I first learned about it from a non-fiction book in which these techniques were used by scientists in the US to locate Ru…

I'm not an expert either, but you might be interested in the book 'Superforecasting' by Tetlock and Gardner - they have done some (IMHO) very interesting research on predictions markets. It might be the kind of thing you're looking to research more of!

Re: Karl Friston: a neuroscientist who might hold the key to true AI

#104

Post by lurquer »

klodolph wrote: post #101
lurquer wrote: post #97

Earlier quoted context omitted.

It suffers from the fallacy of petitio principii, in that it assumes arguendo that consciousness is comprised of neurons (and, as mentioned above, that a neuron is less complex than consciousness.) But it's not stated as an 'argument' in any case so perhaps the term fallacy was out-of-place.

Petitio principii is when the premise assumes the truth of the conclusion, but since there is no argument and no conclusion it's impossible for the statement to suffer from that fallacy.

You're right. Shouldn't have posted that. I just wanted the last word. (Dammit... did it again!)

Re: Karl Friston: a neuroscientist who might hold the key to true AI

#105

Post by nuguy »

ss2003 wrote: post #91
nuguy wrote: post #22

This article is a complete waste. The title implies that it’s about ai but it turns out to be a portrait of a mans life — a pr piece. Not only that but free energy minimization has nothing to do with intelligence other than vaguely describing one of its most obvious and superficial characteristics. —- Ai is the most important issue in the world. True general ai is an existential threat to human kind. The economics of…

I agree with most of what you state, but the prohibition of AI is impossible. How could you stop nations from researching it secretly? How could you stop the Amazons and Baidus?

The only thing that is clear is that something must be thought of and attempted.

Re: Karl Friston: a neuroscientist who might hold the key to true AI

#106

Post by sixdimensional »

Chris_Jay wrote: post #103
sixdimensional wrote: post #96

Earlier quoted context omitted.

Since you mention Bayesian methods, I thought I may randomly ask you - have you come across any good work about applications of subjective Bayesian statistics in AI? I was particularly interested in subjective Bayes theory due to the way it seems to interleave human input with mathematical theory. I first learned about it from a non-fiction book in which these techniques were used by scientists in the US to locate Ru…

I'm not an expert either, but you might be interested in the book 'Superforecasting' by Tetlock and Gardner - they have done some (IMHO) very interesting research on predictions markets. It might be the kind of thing you're looking to research more of!

Thanks for the recommendation - that does look very interesting indeed! I will find myself a copy and have a read. Hacker News book club comes through again!! :)

Re: Karl Friston: a neuroscientist who might hold the key to true AI

#107

Post by snrji »

paraschopra wrote: post #98
snrji wrote: post #85

Earlier quoted context omitted.

Excuse me, another followup question (can't edit on mobile): can you ELI5 how do exploitation and exploration "emerge" naturally instead of the tradeoff being explicitly coded as in RL?

In forumations of FEP, there are two terms: cost and ambiguity. Minimisation of this combined term happens in a Bayesian optimal way. So you don't have to explicitly code weights for exploration and exploitation. Although what you do have to code is prior preferences, and since it is a distribution, you implicitly code the range of those preferences. But once you do that the FEP, algorithm figures out when to collect…

I see. Much more elegant than explicitly coding the trade-off, actually :)

Re: Karl Friston: a neuroscientist who might hold the key to true AI

#108

Post by snrji »

atschantz wrote: post #100
snrji wrote: post #85

Earlier quoted context omitted.

Excuse me, another followup question (can't edit on mobile): can you ELI5 how do exploitation and exploration "emerge" naturally instead of the tradeoff being explicitly coded as in RL?

As a general answer, the theory suggests that organisms maximize a quantity known as model evidence, which is just a way of saying 'how much evidence does some data provide for my model of the world?' There are two complementary ways to maximize this - change your model or change your world. If we now grant that actions also maximize model evidence, then actions can either be conducted to sample data that make the mo…

And the optimization process itself would determine whether updating the model or changing the world is optimal, I guess. Thanks.

Re: Karl Friston: a neuroscientist who might hold the key to true AI

#109

Post by eli_gottlieb »

antidesitter wrote: post #58
eli_gottlieb wrote: post #48

Earlier quoted context omitted.

>True AI will build model from the scratch, and not just learn model complexity. There's no such thing as truly learning "from scratch" -- the No Free Lunch Theorem holds no matter what. What you can do is find a sufficiently large (ex: Turing-complete) hypothesis class, and make simplifying assumptions to allow it to be feasibly learnable (such as regularization or priors).

The No Free Lunch Theorem is irrelevant to the real world [0][1]. It assumes all functions, even those with infinite algorithmic complexity, are equally likely. You should look into algorithmic probability for a better foundation. [0] http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.540.... [1] https://arxiv.org/abs/1111.3846.pdf

On the one hand, yes, the No Free Lunch Theorem seems to intuitively rely on the set-theoretic definition of functions, rather than building on a constructive foundation to hypothesize that functions which are "physically harder", in some sense, are less likely.

On the other hand, algorithmic probability requires first defining a Turing machine, rendering the Solomonoff Measure defined only up to a specific programming language, which can bias it some arbitrary amount. That's on top of the Solomonoff Measure itself being incomputable, and so utterly useless as a foundation for real-world machine learning and computational cognitive science.

I agree that positing a Bayesian prior on functions/programs/causal structures gets you around the No Free Lunch Theorem. The question just then ends up being: what sort of hypothesis space, and what sort of prior, sufficiently resemble the real world (the data-generating process) to allow for learning from a given data set? That's a matter of science.

Re: Karl Friston: a neuroscientist who might hold the key to true AI

#110

Post by eli_gottlieb »

sixdimensional wrote: post #96
ArtWomb wrote: post #78

Most grad-level Deep Learning classes have a week or so devoted to "Approximate Bayes" methods. And it's conceivable future updates to all popular probabilistic programming languages will include "programmable" rather than "fixed-function" inference methods. "Inference Metaprogramming" paper https://people.csail.mit.edu/rinard/paper/pldi18.pdf Latest state-of-the-art research will be presented at upcoming NeuroIPS co…

Since you mention Bayesian methods, I thought I may randomly ask you - have you come across any good work about applications of subjective Bayesian statistics in AI? I was particularly interested in subjective Bayes theory due to the way it seems to interleave human input with mathematical theory. I first learned about it from a non-fiction book in which these techniques were used by scientists in the US to locate Ru…

> Since you mention Bayesian methods, I thought I may randomly ask you - have you come across any good work about applications of subjective Bayesian statistics in AI?

https://www.youtube.com/watch?v=O0MF-r9PsvE

https://arxiv.org/abs/1809.10756 (by my adviser)

https://probprog.cc/ (chaired by my adviser, new)

Post reply on HN