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

AlphaGo Zero: Learning from scratch

deepmind.com

Post reply on HN
Print view View original

131–140 of 324 posts

Re: AlphaGo Zero: Learning from scratch

#131

Post by thefalcon »

feelix wrote: post #62
EvgeniyZh wrote: post #41

Earlier quoted context omitted.

Well, I'd love to see NN solution beating top chess engines. It might also introduce novelty to the game, just as regular engines did

It'd be particularly useful to have a chess bot that can play badly in the same way a human does. The problem with the current chess bots is that they play badly, badly. They choose a terrible random mistake to make every few moves, while some of their other moves are brilliant. They cannot accurately mimic beginner or intermediate level players.

This seems like something DeepMind could create, given the incentive. They were able to train AlphaGo to predict human moves in Go at a very high accuracy (obviously not with AlphaGo Zero, but the inferior human-predictive version is how they determined that AGZ is playing qualitatively differently).

Re: AlphaGo Zero: Learning from scratch

#132

Post by thefalcon »

farresito wrote: post #93
narvind wrote: post #16

How I wish Marvin Minsky would have stayed alive for one more year and seen this. He would have been so happy!

I think he was cryopreserved, so he surely will be surprised once they wake him up in the future, assuming cryonics really works.

I'd certainly be surprised if I ever woke up from being cryopreserved. Which isn't to say that I'd object to the process if I had the disposable income and an understanding/cooperative family support structure, which I do not.

Re: AlphaGo Zero: Learning from scratch

#133

Post by thefalcon »

panic wrote: post #36

Is AlphaGo Zero the first Go program without special code to read ladders? I'm curious how a pure neural net can read them, given how non-local they are.

The concept of locality is nothing but a human weakness in Go, the best AI must read the whole board with every move.

EDIT: From the paper: "Surprisingly, shicho (“ladder” capture sequences that may span the whole board) – one of the first elements of Go knowledge learned by humans – were only understood by AlphaGo Zero much later in training" I'm surprised by the author's use of the word "Surprisingly" here.

Re: AlphaGo Zero: Learning from scratch

#134

Post by espadrine »

shmageggy wrote: post #58
habitue wrote: post #30

Earlier quoted context omitted.

I think the fact that it's no longer using Monte Carlo tree search is a huge step forward in the generalizability of the technique. But go is still - a perfect information game - with a relatively small input size (vs. arbitrary computer vision) - cheap to simulate - discrete action space - deterministic This isn't to take away from the magnitude of the achievement, but the nature of the problem itself makes the resu…

It still uses MCTS as its search algorithm. It no longer uses random rollouts as part of the evaluation, though. (Previously it was rollouts/2 + value_network/2)

Random rollouts are what the MC in MCTS stands for.

Without that, it is simply a tree search.

Excerpt from the paper:

> [AlphaGo Zero] uses a simpler tree search that relies upon this single neural network to evaluate positions and sample moves, without performing any Monte-Carlo rollouts.

Re: AlphaGo Zero: Learning from scratch

#135

Post by gizmo686 »

thefalcon wrote: post #133
panic wrote: post #36

Is AlphaGo Zero the first Go program without special code to read ladders? I'm curious how a pure neural net can read them, given how non-local they are.

The concept of locality is nothing but a human weakness in Go, the best AI must read the whole board with every move. EDIT: From the paper: "Surprisingly, shicho (“ladder” capture sequences that may span the whole board) – one of the first elements of Go knowledge learned by humans – were only understood by AlphaGo Zero much later in training" I'm surprised by the author's use of the word "Surprisingly" here.

>Surprisingly, shicho (‘ladder’ capture sequences that may span the whole board)—one of the first elements of Go knowledge learned by humans—were only understood by AlphaGo Zero much later in training. [0]

[0] https://www.nature.com/nature/journal/v550/n7676/full/nature...

Re: AlphaGo Zero: Learning from scratch

#136

Post by tedunangst »

bradleybuda wrote: post #89
narvind wrote: post #16

How I wish Marvin Minsky would have stayed alive for one more year and seen this. He would have been so happy!

In the days when Sussman was a novice, Minsky once came to him as he sat hacking at the PDP-6. “What are you doing?”, asked Minsky. “I am training a randomly wired neural net to play Tic-Tac-Toe” Sussman replied. “Why is the net wired randomly?”, asked Minsky. “I do not want it to have any preconceptions of how to play”, Sussman said. Minsky then shut his eyes. “Why do you close your eyes?”, Sussman asked his teacher…

So Sussman was right the first time?

Re: AlphaGo Zero: Learning from scratch

#137

Post by mda »

dsp1234 wrote: post #5

Comparing the top player's ELO with Zero's ELO (assuming numbers are accurate, etc): Your rating: 3664 Opponent's rating: 5000 Probability of winning: 0.000456879355457417 So 1 in 2,200 games... ouch

I don't think you can apply this to alphago. I think probability for a human to beat alphago now is zero.

Lee Sedol's single victory is the first and the last.

Re: AlphaGo Zero: Learning from scratch

#138

Post by roenxi »

jasonwatkinspdx wrote: post #100
aeleos wrote: post #52

The fact that they only used self play with no outside input here is really interesting. I wonder if this system produced more new styles of play. While I am not that familiar with Go, I know in some of the other articles they talk about things like Chinese starts that are specific to certain cultures. I wonder if the fact that it had no outside reinforcement made it produce movements that we have already seen that a…

> I wonder if this system produced more new styles of play. One thing Alpha go has told us clearly is that it thinks human players over value the margin of victory vs the probability of victory.

The Go community learned to understand that the margin of victory is meaningless along time ago. The most famous game of Honinbo Dosaku, a famous Go player from the late 1600s, is arguably a game where he gave a handicap to his opponent and lost by one point. Lee Chang-Ho, who was the reigning champion in the late 90s, had a style that consistently tried to win by small margins.

AlphaGo now appears to be better than humans in all aspects of gameplay, and it better at calculating very thin margins of probability that a human cannot. This is not unique to any individual aspect of its gameplay; against humans it can also win by huge margins depending on what mistakes the human makes.

Post reply on HN