Locomotion on D4RL Gym (random-expert)
90.3HalfCheetah ScoreA2PO
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| A2POseeds=5, iteration=last training, score_type=normalized2024.03 | 90.3 | 112.5 | 109.1 | |
| AWACseeds=5, iteration=last training, score_type=normalized2024.03 | 87.3 | 84.7 | 11.7 | |
| Diffusion-QLseeds=5, iteration=last training, score_type=normalized2024.03 | 86.1 | 102 | 56.3 | |
| CQL+AWseeds=5, iteration=last training, score_type=normalized2024.03 | 80.7 | 109.6 | 108.6 | |
| LAPOseeds=5, iteration=last training, score_type=normalized2024.03 | 52.6 | 82.3 | 0.4 | |
| EQLseeds=5, iteration=last training, score_type=normalized2024.03 | 47.4 | 68.6 | 9.1 | |
| TD3+BCseeds=5, iteration=last training, score_type=normalized2024.03 | 43.1 | 78.8 | 7 | |
| IQLseeds=5, iteration=last training, score_type=normalized2024.03 | 28.6 | 58.5 | 90.9 | |
| MOPOseeds=5, iteration=last training, score_type=normalized2024.03 | 18.5 | 17.2 | 4.6 | |
| CQLseeds=5, iteration=last training, score_type=normalized2024.03 | 15 | 7.8 | 0.3 | |
| BCseeds=5, iteration=last training, score_type=normalized2024.03 | 13.7 | 10.1 | 14.7 | |
| BCQseeds=5, iteration=last training, score_type=normalized2024.03 | 2.1 | 8.5 | 0.6 |