Skip to content

Latest commit

 

History

History
12 lines (10 loc) · 624 Bytes

README.md

File metadata and controls

12 lines (10 loc) · 624 Bytes

A3C

The A3C algorithm implemented using Pytorch and heavily inspired by the implementation found at ikostrikov's repo. Built for a Reinforcement Learning course at Oxford University.

Agent playing Breakout Agent playing Pong Agent playing Space Invaders

TODO:

  • Implement command line arguments
  • Implement resume training feature
  • Make training processes exit upon convergence
  • Make testing process exit when training process exits
  • Implement logging (tensorboard/visdom)