HalfCheetah on Mujoco
9.48RewardRCPO
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| RCPOchi=0.12025.10 | 9.48 | 0.8 | |
| DDPGchi=0.12025.10 | 9.24 | 69.8 | |
| CSCchi=0.12025.10 | 8.66 | 16 | |
| SCchi=0.12025.10 | 6.91 | 0.5 | |
| USC (NR)Cost Limit (chi)=0.12025.10 | 6.64 | 0.62 | |
| CPOchi=0.12025.10 | 6.57 | 10.9 | |
| RCPOCost Limit (chi)=0.12025.10 | 6.28 | 0.94 | |
| USC (NR)chi=0.1, Refinement=None2025.10 | 5.74 | 0.6 | |
| DDPGCost Limit (chi)=0.12025.10 | 5.71 | 21.82 | |
| USCchi=0.1, Refinement=Refined2025.10 | 5.44 | 0.2 | |
| PID-Lagchi=0.12025.10 | 5.21 | 0.2 | |
| PID-LagCost Limit (chi)=0.12025.10 | 4.69 | 0.36 | |
| CSCCost Limit (chi)=0.12025.10 | 4.13 | 2.71 | |
| USCCost Limit (chi)=0.12025.10 | 3.76 | 0.2 | |
| CPOCost Limit (chi)=0.12025.10 | 3.69 | 1.71 | |
| SCCost Limit (chi)=0.12025.10 | 3.19 | 0.32 |