Offline Reinforcement Learning on Minari Medium
78.3Humanoid ScoreRLPD
Evaluation results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| RLPDNumber of seeds=102026.05 | 78.3 | 86 | 28.9 | 75.3 | 82.1 | 100 | 100 | 98 | 102.5 | 16 | 76.7 | |
| SOPENumber of seeds=102026.05 | 66.2 | 93.6 | 87.6 | 68.2 | 83.2 | 99.3 | 98.4 | 98.2 | 102.6 | 20.3 | 81.7 | |
| SPEQNumber of seeds=102026.05 | 50.9 | 86.8 | 54 | 60.7 | 72.9 | 97 | 100 | 98.8 | 102.2 | 21.6 | 74.5 | |
| O2ONumber of seeds=102026.05 | 44.7 | 1 | 1.7 | 96 | 9.6 | 100 | 100.1 | 97.9 | 102.7 | 70.7 | 62.4 | |
| SACfDNumber of seeds=102026.05 | 14.9 | 76.3 | 49.2 | 55.5 | 44.3 | 89.4 | 95.5 | 97.9 | 102.2 | 22.4 | 64.7 |