Offline Multitask Reinforcement Learning on Franka Kitchen kitchen-mixed
83Average Episodic ReturnDMM
Evaluation results
| Method | Links | |
|---|---|---|
| DMMArchitecture=SSM-based2026.02 | 83 | |
| DCArchitecture=Transformer-based2026.02 | 71.8 | |
| WTArchitecture=Transformer-based2026.02 | 70.9 | |
| DMArchitecture=SSM-based2026.02 | 59.3 | |
| CQLArchitecture=Value-based2026.02 | 51 | |
| IQLArchitecture=Value-based2026.02 | 51 | |
| DS4Architecture=SSM-based2026.02 | 47.7 | |
| DiSPO2024.03 | 46 | |
| DiSPO2024.03 | 46 | |
| TD3+BCArchitecture=Value-based2026.02 | 45.1 | |
| GC-Oracle2024.03 | 43 | |
| DTArchitecture=Transformer-based2026.02 | 25.8 | |
| QLDTArchitecture=Transformer-based2026.02 | 17.7 | |
| USFA2024.03 | 10 | |
| USFA2024.03 | 10 | |
| FB2024.03 | 5 | |
| FB2024.03 | 5 | |
| RaMP2024.03 | 0 | |
| MOPO2024.03 | 0 | |
| COMBO2024.03 | 0 | |
| RaMP2024.03 | 0 | |
| MOPO2024.03 | 0 | |
| COMBO2024.03 | 0 |