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

AlphaGo beats the world champion Lee Sedol in first of five matches

twitter.com

Post reply on HN
Print view View original

491–500 of 596 posts

Re: AlphaGo beats the world champion Lee Sedol in first of five matches

#491

Post by d0m »

amoonki wrote: post #467
d0m wrote: post #404

Earlier quoted context omitted.

I know I will offend chess players by saying this but.. I feel like chess is more of a IQ contest, while Go is more Art-ish in how you move and slowly cripple and surround your opponent. Realistically speaking, there aren't that many moves you can do in chess. Most of them are just blunder that would get you insta-killed by a good player. Contrast that to Go where there are so many good moves. This is why I think the…

Why wouldn't the same be true for Go? Because the search space of good moves is larger?

Totally could be the same for Go, but yeah the search space seems much bigger. I feel in chess there are a few good moves on every turn whereas for Go it seems there are so many. But then, it may be because I'm less good at Go and I don't see the "obvious" move, not that I necessarily see it in chess but you know what I mean.

Re: AlphaGo beats the world champion Lee Sedol in first of five matches

#492

Post by PeterisP »

tel wrote: post #291
exDM69 wrote: post #172

Earlier quoted context omitted.

1202 CPUs and 176 GPUs is the figure mentioned in the Nature paper. But it's important to understand that this is the computer used to train the networks used by the algorithm. It took about 30+ days worth of wallclock to train it. That's about 110 megawatt-hours (MWh) worth of energy required! During the play, the computational requirements are vastly less (but I don't know the figures). It's still probably more tha…

Are they vastly less, though? The core of the algorithm is still a deep Monte Carlo Tree Search which AlphaGo gets quite a boost on computationally for being able to fire it off in parallel. It's obviously incorrect to take the training system and assume it's identical to the live system, but I think it's disingenuous to say the live system didn't have some serious horsepower.

Yes, for neural networks usually training them takes many orders of magnitude more resources than just using them.

For this particular example, training a system involves (1) analysis of every single game of professional go that has been digitally recorded; and (2) playing probably millions of games "against itself", both of which require far more computing power than just playing a single game.

Re: AlphaGo beats the world champion Lee Sedol in first of five matches

#495

Post by jtolmar »

CamperBob2 wrote: post #470
GolDDranks wrote: post #203

Earlier quoted context omitted.

This. AlphaGo utilizes the "Monte Carlo tree search" as its base algorithm[1]. The algorithm has been used for ten years in Go AIs, and when it was introduced, it made a huge impact. The Go bots got stronger overnight, basically. What novel thing AlphaGo did, was a similar jump in algorithmic goodness. It introduced two neural networks for 1) predicting good moves at the present situation 2) evaluating the "value" of…

Especially 2) has been hard to do in Go, without playing the game 'till the end. This is what struck me as especially interesting, as a non-player watching the commentary. The commentators, a 9-dan pro and the editor of a Go publication, were having real problems figuring out what the score was, or who was ahead. When Lee resigned the game, it came as a total surprise to both of them. Just keeping score in Go appears…

Score in Go is captured stones plus surrounded empty territory at the end of the game. Captures are well defined when they happen, but territory is not defined until the end.

The incentive structure of the game leads to moves that firmly define territory usually being weaker, so the better the players, the more they end up playing games where territory is even harder to evaluate.

Re: AlphaGo beats the world champion Lee Sedol in first of five matches

#496

Post by argonaut »

empath75 wrote: post #278
simonh wrote: post #61

Earlier quoted context omitted.

Highly optimised single-function algorithms like this are impressive stuff and can lead to useful tools, but that's it. This gets us no closer to strong AI than a tic tac toe program. Until we have systems that can tackle a wide range of fundamentally different problems and independently adapt strategies for dealing with one class of problems to deal with other classes of problems, systems like Alphago will remain on…

1) this isn't a single function algorithm 2) the human mind is FULL of ugly, highly optimized hacks that accomplish one thing well enough for us to survive. Don't assume that human intelligence is this magical general intelligence, rather than a collection of single function algorithms.

It is. Value and policy networks are nonlinear approximators for value and policy functions.

You're making the mistake of assuming anything about how the human brain learns.

Re: AlphaGo beats the world champion Lee Sedol in first of five matches

#497

Post by argonaut »

Houshalter wrote: post #461
drawnwren wrote: post #191

Earlier quoted context omitted.

Sorry, you're off base a bit. The Atari system did use a Deep Neural Network / Reinforcement algorithm, but as the original poster was trying to point out, the rules of Go were very much hard coded into AlphaGo. From what this [1] says, multiple DNNs are learning how to traverse Monte Carlo trees of Go games. The reinforcement piece comes in choosing which of the Go players is playing the best games. While the higher…

It's best to say that alphago uses neural networks, which are extremely general. The same way planes and cars both use internal combustion engines. ICEs are extremely general. They produce mechanical energy from gas, and are totally uncaring whether you put them into a plane or a car. The body of the plane is necessary, but isn't really the interesting part. Likewise NNs are uncaring what application you put them int…

Even without the neural net system, AI is able to beat most amateurs, and predict moves experts would make.

Re: AlphaGo beats the world champion Lee Sedol in first of five matches

#498

Post by argonaut »

Houshalter wrote: post #466
argonaut wrote: post #240

Earlier quoted context omitted.

Backpropagation and convolutional neural nets were breakthroughs that were immediately put to use.

The point is that no one could train deep nets 10 years ago. Not just because of computing power, but because of bad initializations, and bad transfer functions, and bad regularization techniques, etc. These things might seem like "small iterative refinements", but they add up to 100x improvement. Even when you don't consider hardware. And you should consider hardware too, it's also a factor in the advancement of AI.…

They could. There was a different set of tricks that didn't work as well (greedy pretraining).

Re: AlphaGo beats the world champion Lee Sedol in first of five matches

#499

Post by argonaut »

PascalsMugger wrote: post #319
argonaut wrote: post #296

Earlier quoted context omitted.

I'm not going to fear an AI whose idea of "the world" is a computer game, an AI that isn't even aware of the existence of the real world, and isn't even aware of the existence of the set of real world actions.

I think, almost by definition, you won't be afraid of anything until it's already coming for you. By that time it will already have the capability to decide your future.

The thing is you are essentially making an unfalsifiable argument, invoking the existence of something that is merely imaginary.

Re: AlphaGo beats the world champion Lee Sedol in first of five matches

#500

Post by argonaut »

ogrisel wrote: post #472
argonaut wrote: post #168

Earlier quoted context omitted.

And? This does not address my argument that the complexity is beyond-combinatorially explosive (infinite spaces). I'm not talking about the space of possible board states. I'm talking about merely the set of all possible actions. EDIT: clarified my language to address below reply.

...and it's possible to train learning agents to sense and interact with a world described by high dimensional continuous vector spaces, for instance using conv nets (for sensing audio / video signals) and actor-critic to learn an continuous policy: http://arxiv.org/abs/1509.02971 The fact that the (reinforcement) learning problem is hard or not is not directly related to whether the observation and action spaces are…

There is a near infinite number of such spaces.
Post reply on HN