Offline Reinforcement Learning on D4RL Walker Medium
77.84RewardTransformer
Evaluation results
| Method | Links | |
|---|---|---|
| Transformer2024.05 | 77.84 | |
| BC2022.02 | 75.3 | |
| Aaren2024.05 | 74.44 | |
| AWAC2022.02 | 72.4 | |
| FLOWFORMERLayers=3, Hidden Channels=256, Heads=42022.02 | 71.7 | |
| cosFormerLayers=3, Hidden Channels=256, Heads=42022.02 | 71.4 | |
| DTLayers=3, Hidden Channels=256, Heads=42022.02 | 70.6 | |
| ReformerLayers=3, Hidden Channels=256, Heads=42022.02 | 65.8 | |
| PerformerLayers=3, Hidden Channels=256, Heads=42022.02 | 63.3 | |
| Linear Trans.Layers=3, Hidden Channels=256, Heads=42022.02 | 57.9 |