Offline Reinforcement Learning on D4RL MuJoCo Halfcheetah medexp standard
108.2Normalized ScoreMOBILE
Evaluation Results
| Method | Links | |
|---|---|---|
| MOBILELearning Approach=Model-Based2024.02 | 108.2 | |
| EDACLearning Approach=Model-Free2024.02 | 106.3 | |
| RAVLLearning Approach=Model-Based2024.02 | 102.1 | |
| RAMBOLearning Approach=Model-Based2024.02 | 95.4 | |
| CQLLearning Approach=Model-Free2024.02 | 95 | |
| MOPOLearning Approach=Model-Based2024.02 | 90.8 | |
| COMBOLearning Approach=Model-Based2024.02 | 90 | |
| BCLearning Approach=Model-Free2024.02 | 44 |