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

Lee Sedol Beats AlphaGo in Game 4

gogameguru.com

Post reply on HN
Print view View original

221–230 of 471 posts

Re: Lee Sedol Beats AlphaGo in Game 4

#221

Post by aninhumer »

denim_chicken wrote: post #190
mikeyouse wrote: post #11

Relevant tweets from Demis; Lee Sedol is playing brilliantly! #AlphaGo thought it was doing well, but got confused on move 87. We are in trouble now... Mistake was on move 79, but #AlphaGo only came to that realisation on around move 87 When I say 'thought' and 'realisation' I just mean the output of #AlphaGo value net. It was around 70% at move 79 and then dived on move 87 Lee Sedol wins game 4!!! Congratulations! H…

It feels really weird to see someone being showered with congratulations for beating a computer program. What exactly is he being congratulated for? For probably triggering and then capitalizing on a bug in AlphaGo's AI? For showing that human resolve, perseverance and a "fighting spirit" can trump a flawed AI, at least until the AI gets fixed? For giving DeepMind extremely valuable test data that will only accelerat…

If developments in AI can delegitimise human intelligence, then it was never "legitimate" in the first place (whatever that means).

Re: Lee Sedol Beats AlphaGo in Game 4

#222

Post by tel »

keypusher wrote: post #40

The crucial play here seems to have been Lee Seedol's "tesuji" at White 78. From what I understand this phrase in Go means something like "clever play" but is something like sneaking up on your opponent with something that they did not see coming. Deepmind CEO confirmed that the machine actually missed the implications of this move as the calculated win percentage did not shift until later. https://twitter.com/demish…

To underscore hyperpape:

Tesuji isn't a trick play, it's more like a power play. Each player can read out how a fight is going and see their line far into the future. Two professionals will pick two lines, two suji, which are in balance and push up against one another tightly.

A tesuji is a part of the line which is suddenly showy or strong. It could mean a failure for the opponent if they had not taken enough of an advantage in the struggle to this point or if they do not have a counter tesuji available.

Indeed, that might be the design of a set line: one side continually loses ground to the other forcing the other to take these small advantages all so that the first side has an opportunity to play a tesuji and return to balance. Many such lines are canonicalized ("joseki") and known to any professional. Moreover, professionals regularly identify potential tesuji and expect their opponents to as well.

Re: Lee Sedol Beats AlphaGo in Game 4

#223

Post by markeroon »

kotach wrote: post #131
argonaut wrote: post #130

Earlier quoted context omitted.

Yes, I'm pretty sure we're not talking about the same thing. I'm precisely talking about the trajectories not seen problem. Nothing is going to save you from the fact that the net has not seen a certain state before.

That's not really a problem. Given a large enough dataset you want to generalize from it - there are always states not present in the dataset - the whole point is now to extract features out of your dataset to allow generalization on unseen states. Seeing all of the Go games isn't possible. The compounding errors problem that stems from decision bias isn't because you haven't seen the trajectory, it is because the mo…

This is great stuff, thank you.

Re: Lee Sedol Beats AlphaGo in Game 4

#224

Post by thomasahle »

panic wrote: post #161
karussell wrote: post #145

Earlier quoted context omitted.

Just in case someone wants the commentary around this move 78 https://www.youtube.com/watch?v=yCALyQRN3hw&t=11413

Here's the Myungwan Kim commentary around that same move: https://www.youtube.com/watch?v=SMqjGNqfU6I&t=1h33m . He had been looking with Hajin Lee at variations involving the move beforehand ( https://www.youtube.com/watch?v=SMqjGNqfU6I&t=1h28m1s ), so he immediately noticed that AlphaGo made a possible error in its response.

But Kim also jumped a bit backwards and forward for a few moves, until he finally decided it was a mistake. It didn't seem 100% clear.

Re: Lee Sedol Beats AlphaGo in Game 4

#225

Post by glasslion »

comboy wrote: post #120
mikeyouse wrote: post #11

Relevant tweets from Demis; Lee Sedol is playing brilliantly! #AlphaGo thought it was doing well, but got confused on move 87. We are in trouble now... Mistake was on move 79, but #AlphaGo only came to that realisation on around move 87 When I say 'thought' and 'realisation' I just mean the output of #AlphaGo value net. It was around 70% at move 79 and then dived on move 87 Lee Sedol wins game 4!!! Congratulations! H…

I'll risk assumption that somebody from Deepmind team is reading this. Guys, please, publish charts of win prob estimated by alpha go in time during these games. Some heatmap telling which moves did it consider as best for both sides during the games would also be cool, but that's surely more time consuming to prepare. It would be great to be able to have such things for top pro tournaments in the future.

Actually, it's a tradition that both Players should replay the game after the match, discuss about good moves, bad moves and what they were thinking during the match.

I felt so sorry for Lee Sedol when I saw him lose the second match, facing an empty chair ,and he could only ask one of his friend to review the game.

https://zh.wikipedia.org/wiki/%E5%A4%8D%E7%9B%98

Re: Lee Sedol Beats AlphaGo in Game 4

#226

Post by nathan_f77 »

kronion wrote: post #48

After AlphaGo won the first three games, I wondered not if the computer had reached and surpassed human mastery, but instead how many orders of magnitude better it was. Given today's result, it may be only one order, or even less. Perhaps the best human players are relatively close to the maximum skill level for go, and that the pros of the future will not be categorically better than Lee Sedol is today.

Yeah, I was waiting for AlphaGo to connect up 10 seemingly unrelated moves into some amazing unpredictable shape that wins the game. But now I think that maybe some humans actually have the ability to play a near-perfect game of Go. Maybe the most skilled human players have already nearly reached the peak of what is possible in a Go game.

Re: Lee Sedol Beats AlphaGo in Game 4

#227

Post by tel »

argonaut wrote: post #92
emcq wrote: post #82

Earlier quoted context omitted.

If the value/policy model is predictive with a dataset containing only amateur games, but fails to generalize to unseen data with professional games, that seems like a case of overfitting to a dataset only containing amateur games. In this case the expected value network may be different for amateur games than professional games. Is there something I'm missing?

Sorry, I'm being a little academic. Overfitting is when the model fits to noise or error. Overfitting is not synonymous with "inability to generalize beyond the train and test distribution." For all we know AlphaGo has perfectly fit amateur games, but professional games are on a whole different level

All depends on how you define your data sets I suppose. IGS games include some professional games which, if it is the case that AG is perfectly trained on the amateur mode, were smoothed away.

Re: Lee Sedol Beats AlphaGo in Game 4

#228

Post by kotach »

argonaut wrote: post #199
kotach wrote: post #131

Earlier quoted context omitted.

That's not really a problem. Given a large enough dataset you want to generalize from it - there are always states not present in the dataset - the whole point is now to extract features out of your dataset to allow generalization on unseen states. Seeing all of the Go games isn't possible. The compounding errors problem that stems from decision bias isn't because you haven't seen the trajectory, it is because the mo…

Yes, Hal Daume is referring to the issue I brought up. I'm not interpreting his comments as referring to issues with training the model jointly - he's referring to exactly what I'm describing - never having even seen the expert make a mistake. The only solution is to generate trajectories more intelligently (which is in line with Daume's comments).

Yes, it is true. In the case of Super Mario he does the learning by simulating level-K BFS from positions that resulted in errors (unseen states) and thus minimizes the regret for the next K moves.

Although, if you checkout his papers, the problems I've talked about, when you have more than enough data and when you know you should be able to generalize well you still can get subpar performance if you don't optimize jointly. AlphaGo model isn't optimizied jointly but its power mostly lies in the extreme representation ability of deep neural networks.

Post reply on HN