Reinforcement Learning on DMControl ball-in-cup-catch (expert)
38.87Avg Normalized ScoreSLAC-off+S2P
Evaluation Results
| Method | Links | |
|---|---|---|
| SLAC-off+S2PBase Algorithm=SLAC-off, S2P Augmentation=true2022.09 | 38.87 | |
| CQL+S2PBase Algorithm=CQL, S2P Augmentation=true2022.09 | 28.62 | |
| IQL+S2PBase Algorithm=IQL, S2P Augmentation=true2022.09 | 28.6 | |
| IQLBase Algorithm=IQL, S2P Augmentation=false2022.09 | 28.57 | |
| SLAC-offBase Algorithm=SLAC-off, S2P Augmentation=false2022.09 | 28.56 | |
| CQLBase Algorithm=CQL, S2P Augmentation=false2022.09 | 28.42 |