Offline Reinforcement Learning on Gym Locomotion (Medium)
66.8HalfCheetah ScoreMoReBRAC
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| MoReBRAC2026.01 | 66.8 | 100.3 | 83.07 | 83.39 | |
| SAC-RND2026.01 | 66.4 | 91.1 | 92.7 | 83.4 | |
| ReBRAC2026.01 | 65.6 | 102 | 82.5 | 83.37 | |
| TD3+BC2026.01 | 54.7 | 60.9 | 77.7 | 64.43 | |
| IQL2026.01 | 50 | 65.2 | 80.7 | 65.3 | |
| DD2023.07 | 49.1 | 79.3 | 82.5 | — | |
| MORE2026.01 | 48.9 | 71.5 | 74.8 | 72.53 | |
| IQL2023.07 | 47.7 | 69.2 | 84.5 | — | |
| CQL2023.07 | 47.1 | 70.1 | 83.5 | — | |
| TT2023.07 | 46.9 | 61.1 | 79 | — | |
| GCPC2023.07 | 44.5 | 68 | 78 | — | |
| RvS-R2023.07 | 44.2 | 65.1 | 78.4 | — | |
| BC2023.07 | 43 | 55.8 | 74.1 | — | |
| DT2023.07 | 42.9 | 67.8 | 76.5 | — |