Hand Manipulation on Adroit door-human
25.9Normalized Avg ScoreBPPO
Evaluation Results
| Method | Links | |
|---|---|---|
| BPPOSize=5k2026.01 | 25.9 | |
| C4Size=5k2026.01 | 12.6 | |
| LNSize=5k2026.01 | 11.8 | |
| EDACSeeds=32022.06 | 10.7 | |
| CQLSize=5k2026.01 | 10.7 | |
| DR3Size=5k2026.01 | 10.4 | |
| CQL2022.06 | 9.9 | |
| CQLSeeds=32022.06 | 9.9 | |
| CQL2022.06 | 9.1 | |
| SORLSize=5k2026.01 | 4.9 | |
| SAC-off2022.06 | 3.9 | |
| RORLSeeds=3, Training Epochs=300, Learning Rate=1e-42022.06 | 3.78 | |
| IQLSize=5k2026.01 | 2.9 | |
| UWAC2022.06 | 2.1 | |
| CABI+BCQ2022.06 | 1.7 | |
| BooT-rBootstrapping scheme=r, Trajectory generation scheme=Teacher-forcing2022.06 | 0.6 | |
| TD3BCSize=5k2026.01 | 0.6 | |
| TSRLSize=5k2026.01 | 0.6 | |
| BC2022.06 | 0.5 | |
| BCSeeds=32022.06 | 0.5 | |
| AWR2022.06 | 0.4 | |
| UWACSeeds=32022.06 | 0.4 | |
| TT2022.06 | 0.1 | |
| BooT-oBootstrapping scheme=o, Trajectory generation scheme=Teacher-forcing2022.06 | 0.1 | |
| PBRLSeeds=32022.06 | 0.1 | |
| BCQ2022.06 | 0 | |
| TD3+BCSeeds=32022.06 | 0 | |
| MOPO2022.06 | -0.2 | |
| A2PRSize=5k2026.01 | -0.2 | |
| BEAR2022.06 | -0.3 | |
| COMBO2022.06 | -0.3 | |
| BEARSeeds=32022.06 | -0.3 | |
| DOGESize=5k2026.01 | -0.8 |