Search found 31 matches

by grahamj
Wed Feb 06, 2019 4:23 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: WIP, Eta - GPGPU ANN based engine, RFC
Replies: 10
Views: 1833

Re: WIP, Eta - GPGPU ANN based engine, RFC

small MLP neural networks [...] * game tree in gpu memory * best node selected via score + UCT formula (visit count based) Srdja Small NNs means the evaluation will not be very good, so you need a large search tree. The UCT method requires storing the whole search tree. How will that fit in memory?
by grahamj
Wed Feb 06, 2019 7:40 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: TensorRT Int8 results on volta
Replies: 4
Views: 1031

Re: TensorRT Int8 results on volta

Interesting.The LC0 developers tried an experiment like that, but the result in terms of ELO was disappointing. There was a suggestion that if the training games were generated using INT8 it would be better, but that wouldn't apply to supervised learning. The interactions between tensor cores, Winog...
by grahamj
Tue Feb 05, 2019 7:23 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: chess programming - predictions for the next 5 years
Replies: 15
Views: 2507

Re: chess programming - predictions for the next 5 years

I had no interest in chess programming until I read the AlphaZero paper. I haven't played chess for decades, and never seriously. I worked in computer vision 1995-2007, and followed the machine learning field at that time. The last ten years I have been working in mathematical biology. 15 months ago...
by grahamj
Sat Feb 02, 2019 9:50 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: So, how many of you are working on neural networks for chess?
Replies: 22
Views: 3072

Re: So, how many of you are working on neural networks for chess?

Its not the NN evaluation that is the problem. Its the monte-carlo tree search, and it cannot be solved with tweaks to the evaluation or the monte-carlo tree search. The solution is to stop using MCTS. It is a provably inferior tree search algorithm. The fact that its well suited to parallelization...
by grahamj
Sun Jan 20, 2019 12:06 pm
Forum: Computer Chess Club: General Topics
Topic: Why Lc0 eval (in cp) is asymmetric against AB engines?
Replies: 16
Views: 3063

Re: Why Lc0 eval (in cp) is asymmetric against AB engines?

I have found a quite different type of assymmetry. I use board positions taken randomly from various sets of games, evaluate them with both SF and LC0 using python-chess, convert to prob(win), and graph. I gave them 150ms each. The graph uses 1000 positions from 1000 games from LC0's Test20 self-pla...
by grahamj
Thu Jan 17, 2019 9:17 pm
Forum: Computer Chess Club: General Topics
Topic: Can somebody explain what makes Leela as strong as Komodo?
Replies: 50
Views: 5375

Re: Can somebody explain what makes Leela as strong as Komodo?

Is it the Hardware speed or some breakthrough programming innovation? So far what is the score between Komodo and Leela ? https://www.youtube.com/watch?v=ymDA7-hdkKA My answer is a bit different than others. Leela (following AlphaZero) has found a way to use a powerful GPU for chess, which is itsel...
by grahamj
Thu Jan 17, 2019 4:21 pm
Forum: Computer Chess Club: General Topics
Topic: Can somebody explain what makes Leela as strong as Komodo?
Replies: 50
Views: 5375

Re: Can somebody explain what makes Leela as strong as Komodo?

... But NN just seem to be an "natural" fit for gpus. Srdja Right and this is the essence. If NNs are such a natural fit for GPUs, why is NVIDIA's NN library so big? The Windows DLL is 280Mb, about 100x the size of the SF and LC0 binaries put together. It appears that NVIDIA have spent a huge amoun...
by grahamj
Thu Jan 17, 2019 2:01 pm
Forum: Computer Chess Club: General Topics
Topic: Can somebody explain what makes Leela as strong as Komodo?
Replies: 50
Views: 5375

Re: Can somebody explain what makes Leela as strong as Komodo?

... But NN just seem to be an "natural" fit for gpus. Srdja Right and this is the essence. If NNs are such a natural fit for GPUs, why is NVIDIA's NN library so big? The Windows DLL is 280Mb, about 100x the size of the SF and LC0 binaries put together. It appears that NVIDIA have spent a huge amoun...
by grahamj
Thu Jan 17, 2019 9:18 am
Forum: Computer Chess Club: General Topics
Topic: Can somebody explain what makes Leela as strong as Komodo?
Replies: 50
Views: 5375

Re: Can somebody explain what makes Leela as strong as Komodo?

Like lkaufman, I think it's mainly a successful exploitation of hardware. I look at the AlphaZero paper in two quite different ways. As a scientific contribution, the important thing is the generality of the algorithm. The importance of the `zero' approach is not to avoid human bias, it is to demons...
by grahamj
Sun Jan 13, 2019 8:07 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Writing bugs
Replies: 66
Views: 7761

Re: Writing bugs

mar wrote:
Sat Jan 12, 2019 7:46 pm
You mean something like this? :)
:)