This is a PyTorch implementation of Advantage Actor Critic (A2C), a synchronous deterministic version of A3C "Asynchronous Methods for Deep Reinforcement Learning". Also see the OpenAI post (section A2C and A3C) for more information.

This implementation is inspired by the OpenAI A2C baseline. It uses the same hyper parameters and the model since they were well tuned for Atari games.

Contibutions

Contributions are very welcome. If you know how to make this code better, don't hesitate to send a pull request.

Usage

python main.py --env-name "PongNoFrameskip-v4"

Results

Coming soon.