Locomotion on D4RL Walker2d medium-replay v2
83.8Offline Normalized ReturnCal-QL
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Cal-QLevaluation_rollouts=10, step_interval=50002026.05 | 83.8 | 94.5 | |
| ROIDseeds=3, evaluation_rollouts=10, step_interval=50002026.05 | 80.5 | 131.58 | |
| PEXevaluation_rollouts=10, step_interval=50002026.05 | 73.6 | 89.3 | |
| IQL+GT2026.06 | 72.6 | — | |
| LiRE2026.06 | 71.3 | — | |
| BOORLevaluation_rollouts=10, step_interval=50002026.05 | 69.1 | 114.4 | |
| ODTevaluation_rollouts=10, step_interval=50002026.05 | 68.9 | 76.8 | |
| Oracle2026.06 | 66.9 | — | |
| VOTP2026.06 | 66.3 | — | |
| SURF2026.06 | 64.9 | — | |
| FTB2026.06 | 62.8 | — | |
| DPPO2026.06 | 35.1 | — | |
| P-IQL2026.06 | 32.4 | — | |
| APPO2026.06 | 12.8 | — | |
| IPL2026.06 | 8.6 | — | |
| CPL2026.06 | 6.6 | — | |
| DACERTime steps=1M, Number of random seeds=52026.05 | — | 90.83 | |
| Diff-QLTime steps=1M, Number of random seeds=52026.05 | — | 87.88 | |
| DiffuserTime steps=1M, Number of random seeds=52026.05 | — | 85.51 | |
| DUALTime steps=1M, Number of random seeds=52026.05 | — | 93.15 | |
| EDISTime steps=1M, Number of random seeds=52026.05 | — | 89.53 | |
| IQLTime steps=1M, Number of random seeds=52026.05 | — | 84.04 | |
| RLPDevaluation_rollouts=10, step_interval=50002026.05 | — | 115 |