wassname/DeepRL

mirror of https://github.com/wassname/DeepRL.git synced 2026-08-22 11:40:47 +08:00

T

Shangtong Zhang d5b455b2dc Add figure for A3C

2017-06-03 09:43:44 -06:00

Add figure for A3C

2017-06-03 09:43:44 -06:00

.gitignore

Optimize replay buffer

2017-05-24 19:45:42 -06:00

async_agent.py

Refactor async agents

2017-06-02 22:15:23 -06:00

atari_wrapper.py

Atari wrapper

2017-05-28 21:46:33 -06:00

bootstrap.py

Fix bug for async agents

2017-05-31 15:43:17 -06:00

dqn_agent.py

Optimize DQN

2017-05-29 23:51:23 -06:00

LICENSE

Initial commit

2017-04-20 13:59:53 -06:00

main.py

Update Readme

2017-06-02 22:33:47 -06:00

network.py

Refactor async agents

2017-06-02 22:15:23 -06:00

policy.py

Pixel atari games with DQN

2017-05-20 22:09:18 -06:00

README.md

Minor style fix

2017-06-02 22:34:30 -06:00

replay.py

Fix a bug of replay

2017-05-26 15:18:45 -06:00

task.py

Refactor async agents

2017-06-02 22:15:23 -06:00

README.md

DeepRL

Highly modularized implementation of popular deep RL algorithms powered by PyTorch

Deep Q-Learning
Asynchronous One-Step Q-Learning
Asynchronous One-Step Sarsa
Asynchronous N-Step Q-Learning
Asynchronous Advantage Actor Critic (A3C)

Tested with both classical control tasks (CartPole) and Atari games.

Usage

Detailed usage can be found in main.py

References