Offline Multi-Agent Reinforcement Learning on Multi-agent MuJoCo Ant (Expert, Medium, Medium-Replay)
2,162.8ReturnFACMAC+B3C
Evaluation results
| Method | Links | |
|---|---|---|
| FACMAC+B3CDataset Quality=expert, Observability=Fully observable2025.01 | 2,162.8 | |
| FACMAC+B3CDataset Quality=medium-expert, Observability=Fully observable2025.01 | 2,077.6 | |
| FACMAC+B3CDataset Quality=medium, Observability=Fully observable2025.01 | 1,516.5 | |
| FACMAC+B3CDataset Quality=medium-replay, Observability=Fully observable2025.01 | 1,259.8 | |
| TD3+B3CDataset Quality=medium-replay, Observability=Fully observable2025.01 | 1,226 |