Reinforcement Learning on MuJoCo Ant (test)
7,586Average RewardOracle TC-M2TD3
Evaluation Results
| Method | Links | |
|---|---|---|
| Oracle TC-M2TD3Adversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 7,586 | |
| Oracle TC-RARLAdversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 7,522 | |
| TC-RARLAdversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 7,124 | |
| DRAdversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 6,975 | |
| Stacked TC-RARLAdversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 6,510 | |
| TC-M2TD3Adversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 6,350 | |
| Stacked TC-M2TD3Adversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 6,269 | |
| Oracle M2TD3Adversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 5,528 | |
| M2TD3Adversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 5,330 | |
| Oracle RARLAdversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 4,550 | |
| RARLAdversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 4,153 | |
| TD3Adversary Type=time-constrained fixed cosine, Radius (L)=0.1, Number of seeds=102024.06 | 4,025 | |
| DRTraining Timesteps=5 x 10^72026.06 | 2,740.69 | |
| GERS2026.06 | 2,565.73 | |
| BasePPOTraining Timesteps=3 x 10^72026.06 | 2,154.64 |