Offline Inverse Reinforcement Learning on D4RL HalfCheetah v2 (medium-expert)
10,812.15Cumulative RewardOffline ML-IRL
Evaluation Results
| Method | Links | |
|---|---|---|
| Offline ML-IRLExpert Demonstrations=1,0002023.02 | 10,812.15 | |
| BCExpert Demonstrations=1,0002023.02 | 2,799.4 | |
| CLAREExpert Demonstrations=1,0002023.02 | 292.95 | |
| ValueDICEExpert Demonstrations=1,0002023.02 | -175.87 |