deep_rl
deep_rl copied to clipboard
qgallouedec
Single-file truly minimal implementation of state-of-the-art reinforcement learning algorithms.
deep_rl
Single-file truly minimal implementation of state-of-the-art reinforcement learning algorithms.
Simplicity is the ultimate sophistication. Leonardo da Vinci
| Algorithm | |
|---|---|
| A2C | :x: |
| ARS | :x: |
| C51 | :heavy_check_mark: |
| DDPG | :heavy_check_mark: |
| Double DQN | :x: |
| DQN | :heavy_check_mark: |
| Dueling DQN | :heavy_check_mark: |
| FQF | :x: |
| IQN | :heavy_check_mark: |
| RAINBOW | :x: |
| Recurrent PPO | :x: |
| REINFORCE | :heavy_check_mark: |
| PER (1) | :heavy_check_mark: |
| PPO | :heavy_check_mark: |
| QR-DQN | :x: |
| SAC | :heavy_check_mark: |
| TD3 | :x: |
| TQC | :x: |
| TRPO | :x: |
(1) with DQN
Credit
Most implementations are taken from vwxyzjn/cleanrl without being stricly equivalent. Other ressources: toshikwa/fqf-iqn-qrdqn.pytorch
Contributions
are welcome.
Metadata
22
Stars
0
Forks
Watchers
Owner
qgallouedec
Metadata
Single-file truly minimal implementation of state-of-the-art reinforcement learning algorithms.