DeepChess? Another deep-learning based chess program

Discussion of chess software programming and technical issues.

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
matthewlai
Posts: 787
Joined: Sun Aug 03, 2014 2:48 am
Location: London, UK
Contact:

DeepChess? Another deep-learning based chess program

Post by matthewlai » Mon Oct 17, 2016 5:43 pm

Has anyone seen this paper?

http://www.cs.tau.ac.il/~wolf/papers/deepchess.pdf

I stumbled upon it today. It's by Omid (Falcon author) et al, about using a comparison-based alpha-beta with a deep network. This is something I really wanted to explore but didn't have a chance to.

They claim it's 60 Elo stronger than Crafty, but like Falcon, I don't think the code (or the binary) has been released.
Disclosure: I work for DeepMind on the AlphaZero project, but everything I say here is personal opinion and does not reflect the views of DeepMind / Alphabet.

mar
Posts: 1909
Joined: Fri Nov 26, 2010 1:00 pm
Full name: Martin Sedlak

Re: DeepChess? Another deep-learning based chess program

Post by mar » Mon Oct 17, 2016 10:07 pm

matthewlai wrote:They claim it's 60 Elo stronger than Crafty
Unfortunately they claim so based on 100 games... :shock:

Still pretty impressive. Would be nice to for people to test the binaries.

bob
Posts: 20358
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: DeepChess? Another deep-learning based chess program

Post by bob » Mon Oct 17, 2016 10:22 pm

matthewlai wrote:Has anyone seen this paper?

http://www.cs.tau.ac.il/~wolf/papers/deepchess.pdf

I stumbled upon it today. It's by Omid (Falcon author) et al, about using a comparison-based alpha-beta with a deep network. This is something I really wanted to explore but didn't have a chance to.

They claim it's 60 Elo stronger than Crafty, but like Falcon, I don't think the code (or the binary) has been released.
They also don't explain the circumstances of the 100 game match vs Crafty. Book? No Book? Etc.

It really is pretty sloppy to give 100 game results today, when most play tens of thousands of games to make tuning decisions...

Rochester
Posts: 55
Joined: Sat Feb 20, 2016 5:11 am

Re: DeepChess? Another deep-learning based chess program

Post by Rochester » Tue Oct 18, 2016 5:07 pm

mar wrote:
matthewlai wrote:They claim it's 60 Elo stronger than Crafty
Unfortunately they claim so based on 100 games... :shock:

Still pretty impressive. Would be nice to for people to test the binaries.
You can say the progression not originate from the inside people. They are thinking the same ways always. Only small change of the same thing. Forever small change. But the revolution come from the outside! The new mind can beat the old programme even when work 50 years on the same thing. No new idea in the first 10 years, then never the new idea in the next years also. Mr Hinton is the hero for the computer science! He already invent this long time ago. Only now the computer have the good speed. First the go, now the chess.

CheckersGuy
Posts: 273
Joined: Wed Aug 24, 2016 7:49 pm

Re: DeepChess? Another deep-learning based chess program

Post by CheckersGuy » Tue Oct 18, 2016 6:18 pm

Rochester wrote:
mar wrote:
matthewlai wrote:They claim it's 60 Elo stronger than Crafty
Unfortunately they claim so based on 100 games... :shock:

Still pretty impressive. Would be nice to for people to test the binaries.
You can say the progression not originate from the inside people. They are thinking the same ways always. Only small change of the same thing. Forever small change. But the revolution come from the outside! The new mind can beat the old programme even when work 50 years on the same thing. No new idea in the first 10 years, then never the new idea in the next years also. Mr Hinton is the hero for the computer science! He already invent this long time ago. Only now the computer have the good speed. First the go, now the chess.
However, go and chess are completely different games. Go is much more strategical than chess. Ther reason researchers turned to neural Networks for solving the game of go was because in go it's hard to create a static Evaluation function that tells you how the game may end. I think it will be quite a while until we see a neural Network based Evaluation function in one of the top engines because they are way too slow and impractable today.

Dann Corbit
Posts: 9289
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: DeepChess? Another deep-learning based chess program

Post by Dann Corbit » Tue Oct 18, 2016 8:27 pm

Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.

kinderchocolate
Posts: 406
Joined: Mon Nov 01, 2010 5:55 am
Full name: Ted Wong
Contact:

Re: DeepChess? Another deep-learning based chess program

Post by kinderchocolate » Fri Oct 28, 2016 12:14 pm

I've read the paper, but I don't quite understand:

Q1: How does the binary output representation create the evaluation function? In Giraffe, the weights for the evaluation are the parameters, but I don't see anything like that mentioned in the paper.

Q2: What's the new alpha-beta algorithm? It looks very much like the classical method.

Q3: Why training a one-hot encoding is more accurate than a binary output?

Q4: How does the model deal with two equivalent moves, for example, two different ways to get a checkmate in 5?

Q5: Has the paper been peer-reviewed?

matthewlai
Posts: 787
Joined: Sun Aug 03, 2014 2:48 am
Location: London, UK
Contact:

Re: DeepChess? Another deep-learning based chess program

Post by matthewlai » Thu Nov 03, 2016 1:31 am

kinderchocolate wrote:I've read the paper, but I don't quite understand:

Q1: How does the binary output representation create the evaluation function? In Giraffe, the weights for the evaluation are the parameters, but I don't see anything like that mentioned in the paper.
They don't have an evaluation function in the traditional sense. Everything is based on comparing pairs of positions.
Q2: What's the new alpha-beta algorithm? It looks very much like the classical method.
They use position comparisons. So alpha and beta are both positions instead of scores. Positions don't have scores.
Q3: Why training a one-hot encoding is more accurate than a binary output?
Do they actually use one-hot? In any case I expect them to be the same. If you only have two outputs, softmax one-hot becomes logistic.
Q4: How does the model deal with two equivalent moves, for example, two different ways to get a checkmate in 5?
My guess is the return value will be arbitrary. It wouldn't really matter.
Q5: Has the paper been peer-reviewed?
I believe so, but the code wasn't released, so the reviewer(s) probably did not have access to the actual program.
Disclosure: I work for DeepMind on the AlphaZero project, but everything I say here is personal opinion and does not reflect the views of DeepMind / Alphabet.

whereagles
Posts: 547
Joined: Thu Nov 13, 2014 11:03 am

Re: DeepChess? Another deep-learning based chess program

Post by whereagles » Thu Nov 03, 2016 6:33 pm

Cool. I'm sending this to a friend who's into neural networks and machine learning stuff.

brianr
Posts: 303
Joined: Thu Mar 09, 2006 2:01 pm

Re: DeepChess? Another deep-learning based chess program

Post by brianr » Wed Mar 08, 2017 12:50 pm

An implementation:

https://github.com/mr-press/DeepChess

As an aside, please consider Keras.
I greatly appreciate all those who share their knowledge (and code).
However, as a machine learning newbie with aspirations of someday being a novice (a bit like Carl Spackler's six year plan[http://www.imdb.com/title/tt0080487/), I hope to see greater use of Keras. Keras eliminates much of the pesky TensorFlow v Theano, Python 2 v 3, CPU v GPU, and Linux v Windows background noise.
Even I can get Keras to run.
Thanks again.

Post Reply