Reinforcement Learning on DMControl finger-spin (random)
30.24Averaged Normalized ScoreSLAC-off
Evaluation Results
| Method | Links | |
|---|---|---|
| SLAC-offBase Algorithm=SLAC-off, S2P Augmentation=false2022.09 | 30.24 | |
| SLAC-off+S2PBase Algorithm=SLAC-off, S2P Augmentation=true2022.09 | 27.62 | |
| IQL+S2PBase Algorithm=IQL, S2P Augmentation=true2022.09 | 0.46 | |
| CQLBase Algorithm=CQL, S2P Augmentation=false2022.09 | -0.01 | |
| CQL+S2PBase Algorithm=CQL, S2P Augmentation=true2022.09 | -0.11 | |
| IQLBase Algorithm=IQL, S2P Augmentation=false2022.09 | -0.17 |