Offline Reinforcement Learning on D4RL Gym Locomotion
46.6HalfCheetah MR ScoreDoctor
Evaluation results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Doctor2025.08 | 46.6 | 98.8 | 86.2 | 48.4 | 85.6 | 81.1 | 95.8 | 113.5 | 113.7 | 769.7 | |
| CQL2025.08 | 46.1 | 96.2 | 73.5 | 47 | 58.5 | 72.9 | 91.2 | 99.5 | 109.9 | 694.8 | |
| TD3+BC2025.08 | 45.6 | 72.6 | 81.8 | 47.2 | 60.3 | 84.2 | 92.5 | 101.2 | 111.5 | 696.9 | |
| IQL2025.08 | 45.4 | 93.4 | 79.5 | 48.1 | 66.2 | 79.3 | 89.6 | 106.7 | 109.6 | 717.8 | |
| MTM2025.08 | 43 | 93.1 | 77.3 | 44.1 | 64.9 | 73.4 | 94.7 | 112.4 | 111.4 | 714.3 | |
| BC2025.08 | 40.2 | 36.6 | 25.3 | 41.8 | 55.7 | 63.9 | 55.1 | 53.2 | 100.8 | 472.6 | |
| DT2025.08 | 39.1 | 84.3 | 66 | 42 | 65.2 | 74.5 | 86.8 | 110.6 | 108.3 | 676.8 |