AlphaGo beats the world champion Lee Sedol in first of five matches • Hacker News Board

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

#521

Post by bobbles » Wed, Mar 09, 2016, 11:25 PM UTC

stymaar wrote: post #255

peteretep wrote: post #246

Earlier quoted context omitted.

> physics and experience in working in semiconductors > without some revolutionary technology I can't even > imagine I suspect (in the nicest possible way) that in a lineup of your imagination (on current assumptions) vs the combined ingenuiety of the human race driven by the hidden hand, the latter wins.

> > Give it ten years and […] > I find this overly optimistic exDM69 never said it's not gonna happen, he just said that it's not going to happen in ten years, and I agree with him. Revolutions never occurs that quickly. To achieve that we don't just need an improvement of the current state of the art, we need a massive change and we don't even know what it's going to look like yet ! This kind of revolution may occur…

The thing is though we could already be 10+ years along the path to that next revolution, it wont start being talked about until its basically here

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

#522

Post by Impl0x » Wed, Mar 09, 2016, 11:26 PM UTC

hitekker wrote: post #434

hosh wrote: post #370

Earlier quoted context omitted.

The 2003 match was a brute force approach. AlphaGo's architecture resembles much closer to how humans think and learn. I initially learned Go to be able to have some chance of an AI. I then had some transformative experiences that coincided with my early kyu learning of basic Go lessons. On of the big lessons in Go is to learn how to let go of something. Taking solace in anything on the Go board is one of the blocks…

Is the Monte Carlo approach specific to Go in terms of A.I. challenges? Or is the Monte Carlo approach gaining traction in other A.I. problems as well? I am tremendously unfamiliar with recent A.I developments.

MC tree search isn't specific to Go, no. It's been used for other games including imperfect information ones like poker. I believe that the main reasons for using MC search is that it does not require an evaluation function, and it acts as an anytime algorithm so you can get a "good enough" answer within arbitrary time constraints.

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

#523

Post by eru » Wed, Mar 09, 2016, 11:58 PM UTC

hosh wrote: post #477

nsajko wrote: post #458

Earlier quoted context omitted.

The son is never going to live in a world that doesn't have deep learning. Except if a big solar flare hits us ;}

Or a bunch of other things. Maybe our civilization collapses from peak oil or something.

At the moment, we seem to be having too much oil.

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

#524

Post by eru » Wed, Mar 09, 2016, 11:59 PM UTC

Florin_Andrei wrote: post #425

hosh wrote: post #392

Earlier quoted context omitted.

And just to emphasize the big point here: The AlphaGo that beat the 2p European champion five months ago was not as strong as the AlphaGo that beat Lee Sedol (9p). I don't think this was just the AlphaGo team throwing more hardware. I think they had been constantly running the self-training during the intervening months so that AlphaGo was improving itself. If that is so, then the big thing here isn't that AlphaGo is…

Correct. It played like a top level human player, pretty evenly matched with Lee Sedol. AlphaGo from yesterday would have wiped the floor with AlphaGo from 6 months ago. Various commentators mentioned how both players, human and synthetic, made a few mistakes. Even I caught a slow move made by the AI. So whether Lee Sedol was at the top of his peformance, or not, is a bit of a debate. But the AI was clearly on the sa…

The AI only cares about winning, not about winning by a huge margin.

The slow move might just mean that this was sufficiently big and safer.

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

#525

Post by eru » Thu, Mar 10, 2016, 12:00 AM UTC

jcoffland wrote: post #489

hosh wrote: post #392

Earlier quoted context omitted.

And just to emphasize the big point here: The AlphaGo that beat the 2p European champion five months ago was not as strong as the AlphaGo that beat Lee Sedol (9p). I don't think this was just the AlphaGo team throwing more hardware. I think they had been constantly running the self-training during the intervening months so that AlphaGo was improving itself. If that is so, then the big thing here isn't that AlphaGo is…

This is misleading. AlphaGo beat a 2p player five months ago. Now it has beaten a 9p player. That tells is nothing about it's improvement in the intervening time. Given only this information, however unlikely, AlphaGo could have actually been stronger before.

AlphaGo lost a few of the informal matches.

Also, the people working on it flat out told the world that today's version of AlphaGo beats October's version literally all the time.

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

#526

Post by areyousure » Thu, Mar 10, 2016, 12:01 AM UTC

ghaff wrote: post #455

areyousure wrote: post #450

Earlier quoted context omitted.

Anything written. I'll be particularly happy with higher "quality" sources -- books, quotations in newspapers, etc. -- but honestly, I'm not that picky and will accept an anonymous comment on a random forum.

"Fotland, an early computer Go innovator, also worked as chief engineer of Hewlett Packard’s PA-RISC processor in the 70s, and tested the system with his Go program. “There’s some kind of mental leap that has to happen to get you past that block, and the programs ran into the same issue. The issue is being able to look at the whole board, not the just the local fights.” Fotland and others tried to figure out how to m…

Thank you for the source. I believe this is a good written example of how conservative estimates were as recently as May of 2014.

Nonetheless, the quoted estimate in the article (mentioned twice, including in the second sentence) is "I think maybe ten years", ie 2024, which while inaccurate is probably "in our lifetimes".

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

#527

Post by tel » Thu, Mar 10, 2016, 12:04 AM UTC

PeterisP wrote: post #492

tel wrote: post #291

Earlier quoted context omitted.

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.

I'm very aware of that. What I'm saying is that AlphaGo is not merely a neural net reporting best moves directly off the forward propagation. There are two nets which essentially act as proposal distributions for an exploration/exploitation tradeoff in the search space of game trees by which AlphaGo reads positions essentially out to the end of the game and ranks them by win rate (this is Monte Carlo Tree Search). The net moves are "nice" (I think they run at like 80% win rate against some other Go AIs? Maybe I'm misremembering) but the real heart of what makes AlphaGo play well is the MCTS which requires some vast resources to execute—live resources.

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

#528

Post by ghaff » Thu, Mar 10, 2016, 12:23 AM UTC

areyousure wrote: post #526

ghaff wrote: post #455

Earlier quoted context omitted.

"Fotland, an early computer Go innovator, also worked as chief engineer of Hewlett Packard’s PA-RISC processor in the 70s, and tested the system with his Go program. “There’s some kind of mental leap that has to happen to get you past that block, and the programs ran into the same issue. The issue is being able to look at the whole board, not the just the local fights.” Fotland and others tried to figure out how to m…

Thank you for the source. I believe this is a good written example of how conservative estimates were as recently as May of 2014. Nonetheless, the quoted estimate in the article (mentioned twice, including in the second sentence) is "I think maybe ten years", ie 2024, which while inaccurate is probably "in our lifetimes".

There are a lot of quotes in that article though. And a number are in the vein of not being sure how they were going to get from where they were to better-than-human. Not my field in any case but I think it's fair to say that there was a lot of skepticism about even the general path going forward even relatively recently.

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

#529

Post by YeGoblynQueenne » Thu, Mar 10, 2016, 12:35 AM UTC

escoz wrote: post #462

YeGoblynQueenne wrote: post #416

Earlier quoted context omitted.

Go is literally infinitely more easy to solve than general intelligence. "Literally" in the sense that Go has a finite number of board states, while a general intelligence must be able to deal with an infinite amount of novel situations, presumably by generalising from previously experienced ones. Infinity is a real problem. When you try to learn from examples, you first need to see "enough" examples of whatever you'…

Yes, go have a finite number of board states, just 2.08168199382×10^170, just a bit over the 10^80 atoms in the universe. https://en.wikipedia.org/wiki/Go_and_mathematics

Yes, but the point is that you 're not going to brute-force this. If it's finite you can hope to approximate it. If it's not finite, you can't even approximate it.

Look- take Monte Carlo methods. You can sample a very big number of events and hope to get some useful information from that. If you sample infinity, though, what do you get? Infinity.

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

#530

Post by gjm11 » Thu, Mar 10, 2016, 12:39 AM UTC

stromgo wrote: post #459

gjm11 wrote: post #402

Earlier quoted context omitted.

Fair comment about very weak players. My impression is that the element of chance goes way down well before you get to high dan level, though. How sure are you that I'm wrong?

Quite sure. 17k players still make many huge mistakes, and at the other extreme, God (say 13d) would win 100% of the time against a 12d player. Given that the win ratio for a one rank difference starts at 50% for extreme beginners and ends at 100% for God, maybe you should be the one explaining why you'd expect a significant plateau at 2/3 instead of a smooth increase.

I agree. In particular, I wasn't arguing for anything special to happen at p=2/3. I was hopeful, though, that p=2/3 might be a tolerable approximation over a reasonable range of skill levels.

Having played a bit with some toy models, I've changed my mind a bit; my guess is that p=2/3 is a reasonable approximation for few-dan and few-kyu amateurs, but that outside, say, the 5k-5d range it's far enough off to make a substantial difference.

So, what does this do to those (anyway fairly bogus) "depth" figures? My crappy toy model suggests that for a 2/3 win probability you need a 3-rank difference around 24k, a 2-rank difference around 12k, a 1-rank difference around 2d, a 0.5-rank difference around 8d. And I estimate God at 15 amateur dan (if Cho Chikun is 9p and needs 4 stones from God then God is 21p; if, handwavily, 9d=3p and one p-step is 1/3 the size of one d-step, then God is 21p = (3+18)p = (9+6)d = 15d). So we need maybe 20 steps from God to 5d, then maybe 10 from there to 5k, then maybe 5 from there to 15k, then maybe 5 from there to 30k. That's 40 steps -- not so very different from what we get just by pretending one rank = one "2/3 win probability" step, as it happens.