New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[FEATURE] Implement self-play for two-player zero-sum games #99

Open

5 tasks

RPegoud opened this issue Jul 8, 2024 · 0 comments

Open

5 tasks

[FEATURE] Implement self-play for two-player zero-sum games #99

RPegoud opened this issue Jul 8, 2024 · 0 comments

Labels

Contributor

RPegoud commented Jul 8, 2024

Description:
Add self-play versions of DQN and PPO for two-player zero-sum games in PGX environments.

Checklist:

Determine how to keep the value estimation consistent (e.g. flip the board or reverse the discount for opponent values)
Add PGX environment configs
Implement self-play for DQN
And for PPO
(optional) If possible, for AlphaZero

RPegoud added the enhancement label

RPegoud changed the title ~~[FEATURE] Implement self-play for DQN and PPO~~ [FEATURE] Implement self-play for two-player zero-sum games

RPegoud mentioned this issue

[FEATURE] Implement self-play for two-player zero-sum games #103

Open

5 tasks

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment