Offline Reinforcement Learning on D4RL MuJoCo walker2d medium-replay v2
87.5Normalized ScoreCOUNT
Evaluation results
| Method | Links | |
|---|---|---|
| COUNTtraining steps=1M, seeds=5, version=v22026.03 | 87.5 | |
| ROMItraining steps=1M, seeds=5, version=v22026.03 | 85.2 | |
| CQLtraining steps=1M, seeds=5, version=v22026.03 | 81.8 | |
| RAMBOtraining steps=1M, seeds=5, version=v22026.03 | 81.2 | |
| MOBILEtraining steps=1M, seeds=5, version=v22026.03 | 75 | |
| IQLtraining steps=1M, seeds=5, version=v22026.03 | 74.8 | |
| MOPOtraining steps=1M, seeds=5, version=v22026.03 | 73.6 |