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

We can’t trust AI systems built on deep learning alone

technologyreview.com

Post reply on HN
Print view View original

61–70 of 133 posts

Re: We can’t trust AI systems built on deep learning alone

#61

Post by antome »

mindgam3 wrote: post #58

> AlphaGo can play very well on a 19x19 board but actually has to be retrained to play on a rectangular board. This right here is the soft underbelly of the entire “machine learning as step towards AGI” hype machine, fueled in no small part by DeepMind and its flashy but misleading demos. Once a human learns chess, you can give it a 10x10 board and she will perform at nearly the same skill level with zero retraining.…

This doesn't have much to do with the algorithms, and is more to do with the engineering decisions that went into AlphaGo and AlphaZero. They are designed to play one combinatorial game really well. With a bit of additional efffort and a lot of additional compute, you could expand the model to account for multiple rule / scale variations, maybe even different combinatorial games.

Re: We can’t trust AI systems built on deep learning alone

#62

Post by milaresearcher »

6gvONxR4sf7o wrote: post #41

I'm coming to suspect that even our data isn't enough for useful AI. Imagine you had a truly general sci-fi AI at your office. It still couldn't just look at your database and answer a simple question like "What was the difference in client churn rates between mobile and desktop last month?" or "What was the effect of experiment 1234 on per-client revenue?" Hell, a human couldn't do it. As far as the human or AI woul…

I totally agree with your statement about AGI, but wouldn't be as pessimistic about neural networks in general. Of course our data is enough for useful deep neural models! Many problems can be solved without it, but in areas of computer vision and speech recognition they seem to be the best (currently known) choice.

Your point about AGI which needs to ask questions about data provenance is super interesting. Are you aware of the line of inquiry into active learning? It's fascinating and has a long history: https://papers.nips.cc/paper/1011-active-learning-with-stati...

https://en.wikipedia.org/wiki/Active_learning_(machine_learn...

Re: We can’t trust AI systems built on deep learning alone

#63

Post by mindgam3 »

jayd16 wrote: post #60
mindgam3 wrote: post #58

> AlphaGo can play very well on a 19x19 board but actually has to be retrained to play on a rectangular board. This right here is the soft underbelly of the entire “machine learning as step towards AGI” hype machine, fueled in no small part by DeepMind and its flashy but misleading demos. Once a human learns chess, you can give it a 10x10 board and she will perform at nearly the same skill level with zero retraining.…

>Once a human learns chess, you can give it a 10x10 board and she will perform at nearly the same skill level with zero retraining. Interesting. Has this actually been shown? I would assume a lot of the strategies a human is familiar with would fall apart as well. I'm no chess or go player but I would have to learn new strategies in a tic-tac-toe game scaled to 10x10. I would certainly not be as proficient although I…

Almost all of the human strategies and concepts would still apply: center control, square control, development, initiative, king safety, the opposition, etc. The only exceptions would be fringe concepts like opening theory (already moot in Chess960) and endgame edge cases.

If you’re still not convinced, I’ll prove that skills transfer by playing bullet against anyone who can make a 10x10 variant playable online.

[Edited to come across less egotistical]

Re: We can’t trust AI systems built on deep learning alone

#64

Post by inimino »

6gvONxR4sf7o wrote: post #41

I'm coming to suspect that even our data isn't enough for useful AI. Imagine you had a truly general sci-fi AI at your office. It still couldn't just look at your database and answer a simple question like "What was the difference in client churn rates between mobile and desktop last month?" or "What was the effect of experiment 1234 on per-client revenue?" Hell, a human couldn't do it. As far as the human or AI woul…

If you don't know the sampling regime that generated the data, you might as well give up. The only solutions to this seem to be collecting and providing all the data to ML analysis, or using provenance to inform the training procedure, which requires human expertise.

However, with a more general AI, we would be able to tell it "this is where and how the data were collected" and it could make the necessary inferences. Fully general AI would also be able to ask the right questions and make reasonable guesses on its own. Everything you do now, and nothing like anything that's been developed.

To me the weak point of this article isn't in the thesis (of course everyone can agree that general intelligence is more useful than narrow, all else being equal) but that there was nothing said about how to get there. The only reason ML is currently resurgent is because it we've figured out how to do something that works, while general intelligence has proven beyond our reach for 60+ years.

Edit: actually it seems to be "classical AI" and hybrid approaches and I guess for more details one would need to read the book.

I guess articles like this are worthwhile to temper expectations of what's possible with the current crop of technologies for those not in the field, which could help prevent another winter due to overinflated expectations.

Re: We can’t trust AI systems built on deep learning alone

#65

Post by mindgam3 »

antome wrote: post #61
mindgam3 wrote: post #58

> AlphaGo can play very well on a 19x19 board but actually has to be retrained to play on a rectangular board. This right here is the soft underbelly of the entire “machine learning as step towards AGI” hype machine, fueled in no small part by DeepMind and its flashy but misleading demos. Once a human learns chess, you can give it a 10x10 board and she will perform at nearly the same skill level with zero retraining.…

This doesn't have much to do with the algorithms, and is more to do with the engineering decisions that went into AlphaGo and AlphaZero. They are designed to play one combinatorial game really well. With a bit of additional efffort and a lot of additional compute, you could expand the model to account for multiple rule / scale variations, maybe even different combinatorial games.

> They are designed to play one combinatorial game really well.

Maybe, but they’re certainly not described that way by whoever is in charge of publishing DeepMind’s research:

“A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play”

https://deepmind.com/research/publications/general-reinforce...

Re: We can’t trust AI systems built on deep learning alone

#66

Post by inimino »

Mathnerd314 wrote: post #57

> General AI also ought to be able to work just as comfortably reasoning about politics as reasoning about medicine. It’s the analogue of what people have; any reasonably bright person can do many, many different things. The average human has extreme difficulty reasoning about politics, while usually being reasonable on medicine (anti-vax being one of many exceptions). And it seems strange to expect a skilled pianist…

> some magic AI that will do everything perfectly

Total straw man. He actually uses an intern as an example in the very next sentence after what you quoted, as you would expect them to be able to read and get up to speed on a new area regardless of what it was. Meanwhile SOTA in NLP is a system that can be built to answer a single kind of question but can't explain why it did so or do anything useful if given an explanation of why its answer was wrong.

Re: We can’t trust AI systems built on deep learning alone

#67

Post by ssivark »

w_t_payne wrote: post #46
uoaei wrote: post #43

Earlier quoted context omitted.

SAT solvers are really fast now. Some sort of "neural SAT problem definition" followed by solving it seems to be an interesting direction, but I'm relatively naive on it all. Not sure how training would work since there's no backprop through Boolean logic.

Differentiable SAT solvers is a thing.

Pointers please; not just off-hand mentions leaving others to scratch their heads about what you mean! :-)

Re: We can’t trust AI systems built on deep learning alone

#68

Post by inimino »

jayd16 wrote: post #33

Has work been done to formally prove general AI can't arise from deep learning? I can't help but feel its an assumption being made by those that prefer classical research.

It is trivially proved that deep learning can represent any computable function so the proof you're asking for is not going to be possible. However, it's also completely obvious to anyone familiar with deep learning as it exists today that it is not moving in the direction of AGI and none of the research is ever going to lead to AGI short of some kind of miracle. The burden of proof here would be on those who think this is possible to give some plausible mechanism (or research agenda) for getting there.

Re: We can’t trust AI systems built on deep learning alone

#69

Post by csande17 »

mindgam3 wrote: post #63
jayd16 wrote: post #60

Earlier quoted context omitted.

>Once a human learns chess, you can give it a 10x10 board and she will perform at nearly the same skill level with zero retraining. Interesting. Has this actually been shown? I would assume a lot of the strategies a human is familiar with would fall apart as well. I'm no chess or go player but I would have to learn new strategies in a tic-tac-toe game scaled to 10x10. I would certainly not be as proficient although I…

Almost all of the human strategies and concepts would still apply: center control, square control, development, initiative, king safety, the opposition, etc. The only exceptions would be fringe concepts like opening theory (already moot in Chess960) and endgame edge cases. If you’re still not convinced, I’ll prove that skills transfer by playing bullet against anyone who can make a 10x10 variant playable online. [Edi…

> make a 10x10 variant playable online

I'm not saying I'm definitely going to do this, but is there a rulebook somewhere for 10x10 chess? (What are the initial piece positions, and how would castling work?)

Re: We can’t trust AI systems built on deep learning alone

#70

Post by mindgam3 »

csande17 wrote: post #69
mindgam3 wrote: post #63

Earlier quoted context omitted.

Almost all of the human strategies and concepts would still apply: center control, square control, development, initiative, king safety, the opposition, etc. The only exceptions would be fringe concepts like opening theory (already moot in Chess960) and endgame edge cases. If you’re still not convinced, I’ll prove that skills transfer by playing bullet against anyone who can make a 10x10 variant playable online. [Edi…

> make a 10x10 variant playable online I'm not saying I'm definitely going to do this, but is there a rulebook somewhere for 10x10 chess? (What are the initial piece positions, and how would castling work?)

I’m not aware of 10x10 chess ever being attempted, let alone codified. Here’s my suggestion: add another set of bishops (or knights, or one of each) in between rooks and king/queen. Castling works the same, king goes around rook.
Post reply on HN