Reinforcement learning Chess experiments (Funding issue)
Sort:
Most Recent
Forum Legend
Following
New Comments
Locked Topic
Pinned Topic
There is an interesting code for reinforcement learning chess available on github at
https://github.com/Zeta36/chess-alpha-zero/tree/master/src/chess_zero
I have tested this code on PC, it is a good working code, although more like an older version of alpha go where policy and value used to have different neural networks.
I am thinking to experiment it on Amazon EC2 on GPUs for a couple of days. Anyone on chess.com who is willing to be a part of the experiment. We might get interesting games like Alpha-zero.
https://aws.amazon.com/ec2/elastic-gpus/pricing/
Do let me know
Regards