Cooperative Multi-Agent Reinforcement Learning on SMAC 8m map
17.8ReturnKD-MARL
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| KD-MARLConfiguration=LH2026.04 | 17.8 | 88.97 | 17.3 | |
| KD-MARLConfiguration=LH+A2026.04 | 17.6 | 88.23 | 15.8 | |
| MAPPOConfiguration=FO2026.04 | 17 | 89.91 | 21.5 | |
| QMIXConfiguration=FO2026.04 | 16 | 92.19 | 21.9 | |
| VDNConfiguration=FO2026.04 | 15 | 75.32 | 19 | |
| MAPPOConfiguration=LH2026.04 | 14 | 77.82 | 22 | |
| QMIXConfiguration=LH2026.04 | 12.5 | 64.78 | 18.1 | |
| MAPPOConfiguration=LH+A2026.04 | 10 | 60.07 | 21.8 | |
| VDNConfiguration=LH2026.04 | 10 | 52.11 | 17.2 | |
| QMIXConfiguration=LH+A2026.04 | 8.5 | 48.13 | 10.8 | |
| VDNConfiguration=LH+A2026.04 | 6 | 33.05 | 15 |