Google's AlphaGo team has been working on chess
Moderators: hgm, Rebel, chrisw
-
- Posts: 7220
- Joined: Mon May 27, 2013 10:31 am
Re: Google's AlphaGo team has been working on chess
Building neural network software from scratch is much work.
-
- Posts: 3657
- Joined: Wed Nov 18, 2015 11:41 am
- Location: hungary
Re: Google's AlphaGo team has been working on chess
[quote="Rein Halbersma"]
[b]Equal budget[/b] would be a fairer comparison since AlphaZero and Stockfish take advantage of different types of hardware (GPU vs CPU).
If you look at the scaling graph of thinking time vs performance, it suggests that Stockfish is still ahead at fast time controls but that at longer time controls AlphaZero dominates. It would be interesting to see this graph as a function of money resources.
[/quote]
I am afraid that Stockfish does not get significant help from a much more expensive hardware than it was used for demonstration of AlphaZero.
Supposing the 64 cores used by them are physical cores and not logical cores the increase of cores number to 128, 256,.. give some ten Elo only.
[b]Equal budget[/b] would be a fairer comparison since AlphaZero and Stockfish take advantage of different types of hardware (GPU vs CPU).
If you look at the scaling graph of thinking time vs performance, it suggests that Stockfish is still ahead at fast time controls but that at longer time controls AlphaZero dominates. It would be interesting to see this graph as a function of money resources.
[/quote]
I am afraid that Stockfish does not get significant help from a much more expensive hardware than it was used for demonstration of AlphaZero.
Supposing the 64 cores used by them are physical cores and not logical cores the increase of cores number to 128, 256,.. give some ten Elo only.
-
- Posts: 859
- Joined: Mon Aug 10, 2009 10:05 pm
- Location: Italy
- Full name: Stefano Gemma
Re: Google's AlphaGo team has been working on chess
I think that this a limit of alphabeta algorithm, not a limit of Stockfish itself. The limit of alphabeta is derived by the nature of the game, of course, that grows exponentially at any ply. A smarter approach that uses neural networks and/or other AI algorithms could potentially gives better performance than alphabeta (as AphaZero seems to "demonstrate"). of course AI requires more computational power than alphabeta based algorithms but could eventually scale better at time/power increasing.corres wrote:I am afraid that Stockfish does not get significant help from a much more expensive hardware than it was used for demonstration of AlphaZero.Rein Halbersma wrote:
Equal budget would be a fairer comparison since AlphaZero and Stockfish take advantage of different types of hardware (GPU vs CPU).
If you look at the scaling graph of thinking time vs performance, it suggests that Stockfish is still ahead at fast time controls but that at longer time controls AlphaZero dominates. It would be interesting to see this graph as a function of money resources.
Supposing the 64 cores used by them are physical cores and not logical cores the increase of cores number to 128, 256,.. give some ten Elo only.
Author of Drago, Raffaela, Freccia, Satana, Sabrina.
http://www.linformatica.com
http://www.linformatica.com
-
- Posts: 273
- Joined: Wed Aug 24, 2016 9:49 pm
Re: Google's AlphaGo team has been working on chess
I think working on a similar engine to AlphaZero would be really intresting. One will obviously not get the same performance as AlphaZero but it would still be intresting to see, how well the algorithm scales (with additional hardware and time) compared to current state of the art engines.
-
- Posts: 536
- Joined: Thu Mar 09, 2006 3:01 pm
Re: Google's AlphaGo team has been working on chess
See here:
https://github.com/Zeta36/chess-alpha-zero
I have gotten it to work, but it is painfully slow with just one gpu.
LeelaZero is trying a more distributed approach like Fishtest, albeit for Go.
https://github.com/gcp/leela-zero
https://github.com/Zeta36/chess-alpha-zero
I have gotten it to work, but it is painfully slow with just one gpu.
LeelaZero is trying a more distributed approach like Fishtest, albeit for Go.
https://github.com/gcp/leela-zero
-
- Posts: 7220
- Joined: Mon May 27, 2013 10:31 am
Re: Google's AlphaGo team has been working on chess
That's fast show us some games.
-
- Posts: 273
- Joined: Wed Aug 24, 2016 9:49 pm
Re: Google's AlphaGo team has been working on chess
That's really nice. The new Titan V from Nvidia would be a really good gpu for training. (10 times the performance of a 1080TI) but it costs 3000$brianr wrote:See here:
https://github.com/Zeta36/chess-alpha-zero
I have gotten it to work, but it is painfully slow with just one gpu.
LeelaZero is trying a more distributed approach like Fishtest, albeit for Go.
https://github.com/gcp/leela-zero
-
- Posts: 5566
- Joined: Tue Feb 28, 2012 11:56 pm
Re: Google's AlphaGo team has been working on chess
Have there been any attempts to use MCTS just for tuning of the eval weights of a "regular" chess engine?
Tuning the eval to predict the outcome of an alpha-beta search is a bit hopeless because of all the tactics that can't be encoded in a typical evaluation function. MCTS might average out the tactics sufficiently that its results can directly be used for tuning. The engine would then use alpha-beta for playing games.
Tuning the eval to predict the outcome of an alpha-beta search is a bit hopeless because of all the tactics that can't be encoded in a typical evaluation function. MCTS might average out the tactics sufficiently that its results can directly be used for tuning. The engine would then use alpha-beta for playing games.
-
- Posts: 273
- Joined: Wed Aug 24, 2016 9:49 pm
Re: Google's AlphaGo team has been working on chess
I dont think that would work particularly well. No matter what optimization algorithm one uses we only have a linear funciton of evaluation parameters which I doubt can encode tactics (very well).syzygy wrote:Have there been any attempts to use MCTS just for tuning of the eval weights of a "regular" chess engine?
Tuning the eval to predict the outcome of an alpha-beta search is a bit hopeless because of all the tactics that can't be encoded in a typical evaluation function. MCTS might average out the tactics sufficiently that its results can directly be used for tuning. The engine would then use alpha-beta for playing games.
Neural networks aren't like that and can even learn non-linear functions well.
As for tuning the "regular" evaluation parameters and dont quite understand what you mean by "using mcts to train eval parameters"
-
- Posts: 5566
- Joined: Tue Feb 28, 2012 11:56 pm
Re: Google's AlphaGo team has been working on chess
Run random simulations to get some sort of "averaged" estimation of the evaluation, the idea being that the tactics of the position are washed out sufficiently by the nature of mcts. Then tune the eval weights to better predict that washed out tactic-free estimation.CheckersGuy wrote:As for tuning the "regular" evaluation parameters and dont quite understand what you mean by "using mcts to train eval parameters"
I don't really expect it to work, but I wouldn't expect AlphaZero Chess to work, either