Action mask for more algorithms

thu-ml / tianshou

An elegant PyTorch deep reinforcement learning library.

https://tianshou.org

MIT License

7.96k stars 1.13k forks source link

Action mask for more algorithms #334

Open Stuhl opened 3 years ago

Stuhl commented 3 years ago

Action masking is present in DQN but not in PPO or other algorithms. Any plans to add those in the future? I assume it's possible to add those with custom policies.

Trinkle23897 commented 3 years ago

Currently we don't have such a plan, but you're welcome if you wish to submit a pull request!