Offline Reinforcement Learning on D4RL Gym walker2d (medium, medium-replay, medium-expert)
105.8Normalized ScoreMoMa QL
Evaluation results
| Method | Links | |
|---|---|---|
| MoMa QL2026.05 | 105.8 | |
| Diffusion-QL2026.05 | 97.5 | |
| IDQL-A2026.05 | 93.4 | |
| X-QL2026.05 | 93 | |
| Consistency-AC2026.05 | 91 | |
| IQL2026.05 | 87.3 | |
| CQL2026.05 | 86.2 | |
| ARQ2026.05 | 85.3 |