Offline Reinforcement Learning on D4RL MuJoCo Hopper standard (mixed)
103.9Normalized ScoreMOBILE
Evaluation results
| Method | Links | |
|---|---|---|
| MOBILELearning Approach=Model-Based2024.02 | 103.9 | |
| MOPOLearning Approach=Model-Based2024.02 | 103.5 | |
| RAVLLearning Approach=Model-Based2024.02 | 103.1 | |
| EDACLearning Approach=Model-Free2024.02 | 101 | |
| RAMBOLearning Approach=Model-Based2024.02 | 99.5 | |
| COMBOLearning Approach=Model-Based2024.02 | 89.5 | |
| CQLLearning Approach=Model-Free2024.02 | 86.3 | |
| BCLearning Approach=Model-Free2024.02 | 16.6 |