Multi-Agent Reinforcement Learning on SMAC corridor (test)
20Average ScoreDDN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DDNFramework=DFAC, Training Length=8M timesteps2021.02 | 20 | — | |
| DIQLTraining Length=8M timesteps2021.02 | 19.68 | — | |
| DMIXFramework=DFAC, Training Length=8M timesteps2021.02 | 19.66 | — | |
| LBIprotocol=offline reinforcement learning2024.10 | 19.5 | — | |
| VDNTraining Length=8M timesteps2021.02 | 19.47 | — | |
| IQLTraining Length=8M timesteps2021.02 | 19.42 | — | |
| OMIGAprotocol=offline reinforcement learning2024.10 | 17.1 | — | |
| ICQprotocol=offline reinforcement learning2024.10 | 16.74 | — | |
| BCQ-MAprotocol=offline reinforcement learning2024.10 | 16.42 | — | |
| QMIXTraining Length=8M timesteps2021.02 | 15.07 | — | |
| OMARprotocol=offline reinforcement learning2024.10 | 8.15 | — | |
| CQL-MAprotocol=offline reinforcement learning2024.10 | 6.64 | — | |
| QMIX2020.03 | — | 100 |