Reinforcement Learning on MuJoCo HalfCheetah (test)
8,174Avg PerformanceOracle TC-M2TD3
Evaluation Results
| Method | Links | |
|---|---|---|
| Oracle TC-M2TD3Adversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 8,174 | |
| Oracle TC-RARLAdversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 7,838 | |
| Stacked TC-M2TD3Adversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 7,173 | |
| TC-M2TD3Adversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 6,797 | |
| DRAdversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 5,490 | |
| Stacked TC-RARLAdversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 5,385 | |
| TC-RARLAdversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 5,109 | |
| M2TD3Adversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 3,634 | |
| Oracle M2TD3Adversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 3,453 | |
| TD3Adversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 2,784 | |
| RARLAdversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 154 | |
| Oracle RARLAdversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | -79 |