Offline Inverse Reinforcement Learning on MuJoCo halfcheetah (medium-replay)
12,174.61Average RewardExpert Performance
Evaluation results
| Method | Links | |
|---|---|---|
| Expert Performanceexpert demonstrations=50002023.02 | 12,174.61 | |
| Offline ML-IRLexpert demonstrations=50002023.02 | 9,236.84 | |
| BCexpert demonstrations=50002023.02 | 4,471.72 | |
| ValueDICEexpert demonstrations=50002023.02 | 1,125.17 | |
| CLAREexpert demonstrations=50002023.02 | 437.18 |