Continuous Control on MuJoCo InvertedDoublePendulum
9,356Average RewardOAC
Evaluation Results
| Method | Links | |
|---|---|---|
| OACTraining steps=1M, Number of trials=102021.02 | 9,356 | |
| SACTraining steps=1M, Number of trials=102021.02 | 9,348 | |
| TOP-TD3Training steps=1M, Number of trials=102021.02 | 9,337 | |
| ND TOP-TD3Training steps=1M, Number of trials=102021.02 | 9,330 | |
| QR-TD3Training steps=1M, Number of trials=102021.02 | 9,299 | |
| TD3Training steps=1M, Number of trials=102021.02 | 8,582 |