Offline Reinforcement Learning on D4RL Medium v0
46.3HalfCheetah ReturnBRAC-v
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| BRAC-vSeeds=3, Normalization=Expert policy = 1002021.06 | 46.3 | 31.1 | 81.1 | — | |
| CQLSeeds=3, Normalization=Expert policy = 1002021.06 | 44.4 | 58 | 79.2 | 26 | |
| BCSeeds=3, Normalization=Expert policy = 1002021.06 | 43.1 | 63.9 | 77.3 | 48.9 | |
| Decision TransformerSeeds=3, Normalization=Expert policy = 1002021.06 | 42.6 | 67.6 | 74 | 51.2 | |
| BEARSeeds=3, Normalization=Expert policy = 1002021.06 | 41.7 | 52.1 | 59.1 | — | |
| AWRSeeds=3, Normalization=Expert policy = 1002021.06 | 37.4 | 35.9 | 17.4 | — |