Offline Reinforcement Learning on MuJoCo HalfCheetah Friction shift
72.13Normalized ScoreCEDGE-Planner
Evaluation Results
| Method | Links | |
|---|---|---|
| CEDGE-PlannerLevel=0.52026.05 | 72.13 | |
| DiffuserLevel=0.52026.05 | 71.25 | |
| IQLLevel=0.52026.05 | 69.8 | |
| MOBODYLevel=0.52026.05 | 69.54 | |
| BOSALevel=0.52026.05 | 68.93 | |
| CEDGE-PolicyLevel=5.02026.05 | 68.56 | |
| DARALevel=0.52026.05 | 64.89 | |
| CEDGE-PolicyLevel=0.12026.05 | 64.31 | |
| MOBODYLevel=5.02026.05 | 59.2 | |
| MOBODYLevel=0.12026.05 | 57.53 | |
| CEDGE-PolicyLevel=0.52026.05 | 52.06 | |
| MOBODYLevel=2.02026.05 | 50.02 | |
| TD3-BCLevel=0.52026.05 | 49.43 | |
| CEDGE-PolicyLevel=2.02026.05 | 49.15 | |
| CEDGE-PlannerLevel=2.02026.05 | 47.93 | |
| DiffuserLevel=2.02026.05 | 46.91 | |
| BOSALevel=2.02026.05 | 46.53 | |
| DARALevel=2.02026.05 | 46.25 | |
| IQLLevel=2.02026.05 | 46.04 | |
| IQLLevel=5.02026.05 | 44.96 | |
| BOSALevel=5.02026.05 | 44.07 | |
| TD3-BCLevel=2.02026.05 | 43.51 | |
| DARALevel=5.02026.05 | 40.06 | |
| TD3-BCLevel=5.02026.05 | 35.83 | |
| CEDGE-PlannerLevel=5.02026.05 | 32.79 | |
| DiffuserLevel=5.02026.05 | 32.11 | |
| IQLLevel=0.12026.05 | 26.39 | |
| DARALevel=0.12026.05 | 23.69 | |
| BOSALevel=0.12026.05 | 12.53 | |
| DiffuserLevel=0.12026.05 | 10.7 | |
| CEDGE-PlannerLevel=0.12026.05 | 10.55 | |
| TD3-BCLevel=0.12026.05 | 8.95 |