Offline Reinforcement Learning on HalfCheetah Medium-Expert D4RL
95.1Normalized ScoreTT
Evaluation results
| Method | Links | |
|---|---|---|
| TTConfiguration=Trifle2023.10 | 95.1 | |
| Trajectory Transformer2024.06 | 95 | |
| TTConfiguration=base2023.10 | 95 | |
| Critic-Guided Decision Transformer2024.06 | 93.6 | |
| Decision Mamba2024.06 | 93.5 | |
| %BC2023.10 | 92.9 | |
| DTConfiguration=Trifle2023.10 | 91.9 | |
| CQL2023.10 | 91.6 | |
| TD3(+BC)2023.10 | 90.7 | |
| Decision Diffuser2024.06 | 90.6 | |
| DD2023.10 | 90.6 | |
| TT(+Q)Configuration=Trifle2023.10 | 89.9 | |
| DTConfiguration=base2023.10 | 86.8 | |
| IQL2023.10 | 86.7 | |
| TT(+Q)Configuration=base2023.10 | 82.3 | |
| TrifleConstraint=Action-Space-Constrained (A=0.5)2023.10 | 81.9 | |
| Diffuser2024.06 | 79.8 | |
| TTConstraint=Action-Space-Constrained (A=0.5)2023.10 | 77.8 |