Offline Reinforcement Learning on risk-sensitive D4RL Hopper Mixed
189CVaR 0.1CQL
Evaluation results
| Method | Links | ||
|---|---|---|---|
| CQL2026.06 | 189 | -21 | |
| OWCPG2026.06 | 372 | 442 | |
| ORAAC2026.06 | 876 | 525 | |
| Diffusion-QL2026.06 | 1,021 | 1,010 | |
| CODAC2026.06 | 1,551 | 1,450 | |
| UDAC2026.06 | 1,620 | 1,523 | |
| RS-Diffuser2026.06 | 1,702 | 1,609 |