Locomotion on D4RL Hopper-medium-replay v2
111.1Online Normalized ReturnBOORL
Evaluation results
| Method | Links | ||
|---|---|---|---|
| BOORLevaluation_rollouts=10, step_interval=50002026.05 | 111.1 | 75.5 | |
| DACERTime steps=1M, Number of random seeds=52026.05 | 110.06 | — | |
| DUALTime steps=1M, Number of random seeds=52026.05 | 109.21 | — | |
| Diff-QLTime steps=1M, Number of random seeds=52026.05 | 108.27 | — | |
| ROIDseeds=3, evaluation_rollouts=10, step_interval=50002026.05 | 106.33 | 56.47 | |
| Cal-QLevaluation_rollouts=10, step_interval=50002026.05 | 106.1 | 95.4 | |
| EDISTime steps=1M, Number of random seeds=52026.05 | 103.55 | — | |
| PEXevaluation_rollouts=10, step_interval=50002026.05 | 103.3 | 89.8 | |
| IQLTime steps=1M, Number of random seeds=52026.05 | 101.01 | — | |
| DiffuserTime steps=1M, Number of random seeds=52026.05 | 96.57 | — | |
| ODTevaluation_rollouts=10, step_interval=50002026.05 | 88.8 | 86.6 | |
| RLPDevaluation_rollouts=10, step_interval=50002026.05 | 58.9 | — |