Multi-Agent Reinforcement Learning on REF rdist
-43Mean Episodic RewardMADDPG
Evaluation results
| Method | Links | |
|---|---|---|
| MADDPGq=22022.10 | -43 | |
| DRE-MARLq=22022.10 | -57 | |
| p2p-MARLq=22022.10 | -63 | |
| MAACq=22022.10 | -72 | |
| MAPPOq=22022.10 | -86 | |
| IQLq=22022.10 | -226 | |
| QMIXq=22022.10 | -280 | |
| DRE-MARLq=72022.10 | -752 | |
| p2p-MARLq=72022.10 | -800 | |
| MAACq=72022.10 | -1,070 | |
| MADDPGq=72022.10 | -1,182 | |
| p2p-MARLq=102022.10 | -1,570 | |
| DRE-MARLq=102022.10 | -1,676 | |
| QMIXq=72022.10 | -1,894 | |
| MAPPOq=72022.10 | -1,947 | |
| IQLq=72022.10 | -2,514 | |
| MAACq=102022.10 | -2,558 | |
| MADDPGq=102022.10 | -2,968 | |
| QMIXq=102022.10 | -3,584 | |
| IQLq=102022.10 | -4,090 | |
| MAPPOq=102022.10 | -5,580 |