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

Is There a Smarter Path to Artificial Intelligence? Some Experts Hope So

nytimes.com

Post reply on HN
Print view View original

71–80 of 98 posts

Re: Is There a Smarter Path to Artificial Intelligence? Some Experts Hope So

#71

Post by garmaine »

gaius wrote: post #17
IshKebab wrote: post #13

Earlier quoted context omitted.

There is something deep there - the neural networks are deep. They have lots of layers. It's in contrast to shallow neural networks. It doesn't mean "deep and meaningful".

It doesn't mean "deep and meaningful". It would not be nearly as well funded were it not for this misconception on the part of budget holders, and everyone involved in ML knows it.

I have never once encountered this misconception until now. Every explanation of deep learning I’ve seen begins with an explanation of the origin of the phrase.

Re: Is There a Smarter Path to Artificial Intelligence? Some Experts Hope So

#72

Post by garmaine »

chii wrote: post #31
choxi wrote: post #28

Earlier quoted context omitted.

> it all can't be deep learning unfortunately, because that's not how we learn If you could humor me, why can't it all be deep learning? It seems like a robust model for how we learn: we start with some priors (random weights), experience something that either confirms or negates those priors (sampling), and then adjust our priors based on that experience. Of course there are a wide variety of architectures and the h…

> we start with some priors (random weights) i dont think we start with random priors. We start with a set of priors that are geared for certain things - facial recognition, language etc. The environment then fine tunes them.

There is no way that the genome encodes synaptic layouts to such a detail as to specify an algorithmic prior. At best it specifies, loosely, the generic architecture of how many neurons, how many layers, how tightly folded, and where the inputs (senses) connect. We develop the same algorithms because we all start with the same priors in largely the same gestational and early infant environments.

Re: Is There a Smarter Path to Artificial Intelligence? Some Experts Hope So

#73

Post by XuMiao »

EGreg wrote: post #65
jonathanstrange wrote: post #6

Maybe good AI requires a combination of symbolic approaches for the knowledge representation, deep parsing for NLP and semantics/discourse models, and machine learning for pattern recognition tasks of any kind. The currently trendy "AI" looks more like massive data mining with powerful ML to me, that's very good for certain tasks but brings us nowhere near real AI. The knowledge representation problem has not yet bee…

Let me suggest something that will turn everything upside down... How much of our “logical reasoning” about the world is just really strong correlations? Look, AlphaGo Zero beat the best chess programs by just precomputing MCTS. What does that say about finding the optimal solution to a problem? It seems MCTS beats even AlphaBeta search, by far. So when you talk about applying rules, you’re using a system that tried…

In general, I agree that for avery complicated task that a machine intelligence performs far better than any human being, human-like reasoning might be just impossible. The logical formula might be too long to be reasonable to any human.

But from another side, current AI has a hard time to perform ordinary human reasoning at all. Those reasoning mostly contains a few logical clauses. Why?

Human like logic usually hide the common sense or context in other words. There is no absolute logic like FOL when we talk about human like logic. The traditional rule based AI system fails because of that.

On the other hand, deep learning are very strong at representing these hidden common sense and context. Look at the word2vec.

We now are at a point to combine these two approaches together not to build AlphaGo, but enable ordinary human like reasoning.

At least that is my hope.

Re: Is There a Smarter Path to Artificial Intelligence? Some Experts Hope So

#74

Post by randcraw »

garmaine wrote: post #68
randcraw wrote: post #64

Earlier quoted context omitted.

Lexical and computational semantics may be superior to FOL, (for machine translation, at least) but they've been around now for a very long time without showing much promise toward thawing past AI Winters. I've not heard anyone propose their adoption as the missing next step toward modeling general knowledge across all AI problem spaces. In fact, populating a LS/CS knowledgebase with requisite facts was a game stoppe…

I’m not sure what you are saying. Computational semantics parses sentences to possible interpretations in first order, or higher order logics. It is not some sort of alternative to FOL. And it is still a rapidly developing field.

I'm trying to suggest (perhaps badly) that formal models of logic modeling like LS / CS have an innate achilles heel (from what I've seen): either A) the knowledgebase is precise but tiny, and will remain so because it had to be meticulously hand crafted to comply with a formal semantic model, or B) the KB is representationally sloppy but big and scaleable, because it was populated automatically and informally (probabilistically), thereby forever limiting its amenability to precise models of reasoning. To wit, you may have your cake, but the morsel is so tiny you'll starve.

I think this tradeoff in quality vs quantity has been central to the inability of most Good Old Fashioned AI (GOFAI) formal symbolic methods to 1) scale up beyond toy academic clean-room problems (from the admittedly little I've seen of commercial LS/CS), as well as 2) support the flexibility and imprecision needed to deal with a messy ill-defined very big world with a seemingly unknowable number of unknowns.

It may be that this venerable yin v yang of knowledge modeling will forever bedevil AI. I don't have the sense that any techniques in the past 30 years have really made headway in breaking this log jam. Deep nets have just tilted the playing field in its favor, so long as we're willing to ignore fussy details like dependence, FOL, and causality.

Re: Is There a Smarter Path to Artificial Intelligence? Some Experts Hope So

#75

Post by gaius »

garmaine wrote: post #70
gaius wrote: post #16

Earlier quoted context omitted.

Lets rename "Deep learning" to "Statistical Compression" The previous name for what is now called “machine learning” was predictive statistics. But a better name for deep learning is “machine intuition” since it seems to work pretty well but can’t be readily explained

> since it seems to work pretty well but can’t be readily explained That is becoming less and less true.

We are a long, long way from cracking open a neural net cat classifier and pointing at the exact equation for the whisker coefficient or the nose cuteness factor.

Re: Is There a Smarter Path to Artificial Intelligence? Some Experts Hope So

#76

Post by alexgmcm »

I think it's obvious to anyone with experience in the field that Deep Learning etc. isn't a path to AGI nor an emulation of biological intelligence.

But it is one thing to know that the current path isn't leading us there and another thing entirely to know which path will.

After all it isn't even clear that emulating biological intelligence would be a wise approach - we didn't build the jumbo jet by copying the birds.

Re: Is There a Smarter Path to Artificial Intelligence? Some Experts Hope So

#77

Post by alexgmcm »

vinchuco wrote: post #8

Hardware enables software

Yeah - I remember reading about memristor based technology that aimed to better emulate neurons in the hardware back when I was in Uni.

That was almost 10 years ago though and I haven't heard much of it since.

Re: Is There a Smarter Path to Artificial Intelligence? Some Experts Hope So

#78

Post by YeGoblynQueenne »

randcraw wrote: post #67
YeGoblynQueenne wrote: post #12

Earlier quoted context omitted.

This is a valid question, I don't know why you (originally) got downvoted. The answer I think, for the applications in industry at least, is that PRISM and other probabilistic programming langauges are not as developed as Prolog, a language that has been around for a good four decades now. Swi-Prolog in particular, is a free and open-source Prolog interpreter with an IDE, a graphical package and a veritable somrgasbo…

But its not clear to me that encoding productions in source code adds any value over encoding them in the various other forms used by expert systems of old. Prolog also compels binary constraint satisfaction, recursive descent parsing, backtracking, and depth-first resolution — all undesirable constraints that are easily avoided using other fact representations or resolution engines. Until the startup in question can…

I wouldn't say that backtracking and depth first search (not resolution) are undesirable. Far as I'm concerned they're pragmatic choices that minimise the amount of resources necessary to perform resolution theorem-proving.

Re: Is There a Smarter Path to Artificial Intelligence? Some Experts Hope So

#79

Post by EGreg »

randcraw wrote: post #66
EGreg wrote: post #65

Earlier quoted context omitted.

Let me suggest something that will turn everything upside down... How much of our “logical reasoning” about the world is just really strong correlations? Look, AlphaGo Zero beat the best chess programs by just precomputing MCTS. What does that say about finding the optimal solution to a problem? It seems MCTS beats even AlphaBeta search, by far. So when you talk about applying rules, you’re using a system that tried…

That's like saying that causation doesn't matter; correlation will suffice. But the causal directionality of separating subject from object in a relation is essential to forming or sharing an idea usefully. In the synthetic game worlds you mention, causality is decided by play sequence. But at no point does the agent ever have to employ abstraction, or 'think about thinking', which diminishes the needed depth of thou…

Funny that you say that. David Hume said that you can’t meaningfully prove causation, only describe things and speak about about correlation.

Also time is a local phenemenon. You can’t simply compare A and B time if A and B are really far apart. When you get to really small scales you would be hard pressed to find which event happened before the other, so you get uncaused events like Virtual Particles in quantum mechanics.

And when you have tons of complexity then what do you mean by cause and effect instead of correlation?

https://www.quantamagazine.org/omnigenic-model-suggests-that...

Re: Is There a Smarter Path to Artificial Intelligence? Some Experts Hope So

#80

Post by rwnspace »

YeGoblynQueenne wrote: post #11
jonathanstrange wrote: post #6

Maybe good AI requires a combination of symbolic approaches for the knowledge representation, deep parsing for NLP and semantics/discourse models, and machine learning for pattern recognition tasks of any kind. The currently trendy "AI" looks more like massive data mining with powerful ML to me, that's very good for certain tasks but brings us nowhere near real AI. The knowledge representation problem has not yet bee…

>> Generally speaking, logical modelling is too limited for good concept representations In principle, First Order Logic and equivalent languages can represent anything that can be represented in natural language. The problem is that in practice it is very hard to transfer into a logic language all the knowledge you might need for a useful system. In fact, this was one motivation for at least one branch of machine le…

>In principle, First Order Logic and equivalent languages can represent anything that can be represented in natural language

If you include a semantics for supervaluationism (Sorites paradox, Keefe on vagueness)!

Thinking of it, I'd love to train a NN against human responses to '"is this a heap, yes/no?" +/-1 grain, repeat'.

Post reply on HN