Multi-Agent Reinforcement Learning on REF rac-dist
6,340Mean Episodic RewardDRE-MARL
Evaluation results
| Method | Links | |
|---|---|---|
| DRE-MARLq=102022.10 | 6,340 | |
| MAPPOq=102022.10 | 5,301 | |
| QMIXq=102022.10 | 5,229 | |
| MADDPGq=102022.10 | 4,981 | |
| DRE-MARLq=72022.10 | 3,323 | |
| MAPPOq=72022.10 | 2,949 | |
| p2p-MARLq=102022.10 | 2,884 | |
| QMIXq=72022.10 | 2,824 | |
| MADDPGq=72022.10 | 2,824 | |
| p2p-MARLq=72022.10 | 1,886 | |
| MADDPGq=22022.10 | 287 | |
| DRE-MARLq=22022.10 | 283 | |
| QMIXq=22022.10 | 255 | |
| MAPPOq=22022.10 | 247 | |
| MAACq=72022.10 | 215 | |
| p2p-MARLq=22022.10 | 213 | |
| IQLq=102022.10 | 181 | |
| MAACq=22022.10 | 106 | |
| IQLq=72022.10 | -26 | |
| IQLq=22022.10 | -159 | |
| MAACq=102022.10 | -230 |