Search found 411 matches

by Rémi Coulom
Tue Nov 14, 2017 1:40 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Neural networks for chess position evaluation- request
Replies: 23
Views: 6098

Re: Neural networks for chess position evaluation- request

By MCTS, I mean the AlphaGo zero approach, where there is no random playouts. Playouts are completely replaced by neural-network evaluation. I believe a large network can read a lot of tactics. The raw neural-network of AlphaGo is (weak) pro strength by itself, with no search at all. Of course, conv...
by Rémi Coulom
Mon Nov 13, 2017 11:49 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Neural networks for chess position evaluation- request
Replies: 23
Views: 6098

Re: Neural networks for chess position evaluation- request

Another funny idea: it might be possible to make a neural network using bitboards in a binarized neural network: https://arxiv.org/abs/1602.02830
by Rémi Coulom
Mon Nov 13, 2017 11:40 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Neural networks for chess position evaluation- request
Replies: 23
Views: 6098

Re: Neural networks for chess position evaluation- request

I am convinced that a convolutional neural network should be able to learn an evaluation function that is considerably better than any hand-made evaluation. But it will be much slower. Whether it can be competitive with traditional engines is difficult to predict. It is fun and exciting to try, anyw...
by Rémi Coulom
Mon Nov 13, 2017 7:15 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Neural networks for chess position evaluation- request
Replies: 23
Views: 6098

Re: Neural networks for chess position evaluation- request

I took a longer look at what you did. Max pooling is used for image recognition, but makes little sense for chess. You'll need several layers of convolutions before reducing the resolution. I would simply stack a dozen 3x3 convolutions, and then directly go to a couple scalar layers, without any max...
by Rémi Coulom
Mon Nov 13, 2017 7:08 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Neural networks for chess position evaluation- request
Replies: 23
Views: 6098

Re: Neural networks for chess position evaluation- request

My advice: 1. Using the final result of the game is fine. This is what we do in Go. You'll need really many games, but it is easy to generate many. 2. I took a quick look at your NN architecture. My advice is: - Use a separate channel for black and white pieces - Instead of black/white, encode input...
by Rémi Coulom
Thu Jun 23, 2016 9:25 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: The right database for client-server web applications? (OT)
Replies: 11
Views: 2514

Re: The right database for client-server web applications? (

Yes, I already have the consolidation operation. It is really very slow. But for my applications, it is not necessary, so I don't worry to much about this. Transactions are not so difficult with a journal. I don't wish to implement transactions that can be rolled back. I don't think they are necessa...
by Rémi Coulom
Thu Jun 23, 2016 7:52 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: The right database for client-server web applications? (OT)
Replies: 11
Views: 2514

Re: The right database for client-server web applications? (

Thanks a lot for your feedback, Stefano. I have just started to develop a joedb server, but it is not functional yet. The journal-only approach makes it relatively simple to do. I hope I'll finish it during the summer. I am quite busy with other projects, so progress is a bit slow. My server will no...
by Rémi Coulom
Thu Jun 23, 2016 7:03 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: The right database for client-server web applications? (OT)
Replies: 11
Views: 2514

Re: The right database for client-server web applications? (

Hi, I take the opportunity of this thread to advertise my joedb: http://www.remi-coulom.fr/joedb/intro.html I have found it frustrating to use an SQL API inside C++. It always produces ugly code. So I decided to develop my own database. It is based on an original journal-centered architecture. I bel...
by Rémi Coulom
Mon Mar 28, 2016 3:47 am
Forum: Computer Chess Club: General Topics
Topic: Lee Sedol vs. AlphaGo [link to live feed]
Replies: 121
Views: 30512

Re: Lee Sedol vs. AlphaGo [link to live feed]

I asked Zen's author (Yamato), one day ago whether Zen uses deep learning and his answer was no. So it reaches 7 dan level without any deep learning. There must be a misunderstanding. Zen's improvement comes from neural networks, too. http://computer-go.org/pipermail/computer-go/2016-January/008541...
by Rémi Coulom
Wed Mar 23, 2016 4:25 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Stockfish's tuning method
Replies: 43
Views: 24132

Re: Stockfish's tuning method

All my experiments with RSPSA gave terrible results (you can see the plots in my CLOP paper). I believe that plain SPSA is much better.