Multi-Agent Reinforcement Learning on CN rdist
-161Mean Episodic RewardDRE-MARL
Evaluation results
| Method | Links | |
|---|---|---|
| DRE-MARLq=32022.10 | -161 | |
| p2p-MARLq=32022.10 | -186 | |
| MADDPGq=32022.10 | -213 | |
| MAACq=32022.10 | -270 | |
| IQLq=32022.10 | -552 | |
| MAPPOq=32022.10 | -773 | |
| QMIXq=32022.10 | -1,263 | |
| DRE-MARLq=72022.10 | -3,200 | |
| p2p-MARLq=72022.10 | -3,262 | |
| MAACq=72022.10 | -3,558 | |
| MADDPGq=72022.10 | -3,617 | |
| IQLq=72022.10 | -6,141 | |
| MAPPOq=72022.10 | -7,481 | |
| DRE-MARLq=102022.10 | -8,605 | |
| p2p-MARLq=102022.10 | -8,792 | |
| MAACq=102022.10 | -8,986 | |
| MADDPGq=102022.10 | -9,365 | |
| QMIXq=72022.10 | -10,678 | |
| IQLq=102022.10 | -12,693 | |
| MAPPOq=102022.10 | -19,696 | |
| QMIXq=102022.10 | -22,699 |