Locomotion on D4RL hopper v2 (medium)
103Normalized ReturnCPQL
Evaluation Results
| Method | Links | |
|---|---|---|
| CPQLGradient steps=1M2026.05 | 103 | |
| QDQGradient steps=1M2026.05 | 102.4 | |
| EDACGradient steps=3M2026.05 | 101.6 | |
| EDACGradient steps=3M2026.05 | 101 | |
| PBRLGradient steps=1M2026.05 | 100.7 | |
| CPQLGradient steps=1M2026.05 | 99.7 | |
| QDQGradient steps=1M2026.05 | 99 | |
| SPAARSEnvironment steps=300k, Number of seeds=32026.03 | 92.7 | |
| TTGradient steps=1M2026.05 | 91.5 | |
| DTGradient steps=1M2026.05 | 82.7 | |
| PBRLGradient steps=1M2026.05 | 81.6 | |
| DTGradient steps=1M2026.05 | 67.6 | |
| IQLEnvironment steps=300k, Number of seeds=32026.03 | 66.3 | |
| TTGradient steps=1M2026.05 | 61.1 | |
| UWACGradient steps=1M2026.05 | 50.9 | |
| UWACGradient steps=1M2026.05 | 25.3 |