Offline Reinforcement Learning on Maze2D umaze v1
52.2Normalized ReturnPrivORL-j
Evaluation results
| Method | Links | |
|---|---|---|
| PrivORL-jRL Algorithm=EDAC, epsilon=102025.12 | 52.2 | |
| PrivORL-jRL Algorithm=TD3PLUSBC, epsilon=102025.12 | 49.9 | |
| PrivORL-jRL Algorithm=IQL, epsilon=102025.12 | 49.8 | |
| DP-TransformerRL Algorithm=TD3PLUSBC, epsilon=102025.12 | 49.3 | |
| PrivORL-jRL Algorithm=EDAC, epsilon=12025.12 | 45.5 | |
| PrivORL-jRL Algorithm=IQL, epsilon=12025.12 | 42.1 | |
| DP-TransformerRL Algorithm=IQL, epsilon=102025.12 | 41.2 | |
| DP-TransformerRL Algorithm=EDAC, epsilon=102025.12 | 39.8 | |
| PrivORL-j-URL Algorithm=TD3PLUSBC, epsilon=102025.12 | 38.8 | |
| PrivORL-jRL Algorithm=TD3PLUSBC, epsilon=12025.12 | 38.7 | |
| PrivORL-j-URL Algorithm=EDAC, epsilon=102025.12 | 32 | |
| PrivORL-j-URL Algorithm=IQL, epsilon=102025.12 | 28.9 | |
| DP-TransformerRL Algorithm=EDAC, epsilon=12025.12 | 28.4 | |
| DP-TransformerRL Algorithm=TD3PLUSBC, epsilon=12025.12 | 28.2 | |
| PrivORL-j-URL Algorithm=TD3PLUSBC, epsilon=12025.12 | 26.8 | |
| DP-TransformerRL Algorithm=IQL, epsilon=12025.12 | 25.9 | |
| PrivORL-j-URL Algorithm=IQL, epsilon=12025.12 | 24.6 | |
| PrivORL-j-URL Algorithm=EDAC, epsilon=12025.12 | 11.2 |