Multi-Agent Reinforcement Learning on SMAC Ultra Hard (test) benchmark leaderboard · SOTA2 Research

Multi-Agent Reinforcement Learning on SMAC Ultra Hard (test)

63.35MMM2 7m2M1M vs 8m4M1M ResultDMIX

Evaluation Results

Method	Links
DMIX2023.06		63.35	0	83.52	92.33	81.82	0
DDN2023.06		56.82	0.28	89.77	90.31	67.9	41.19
DPLEX2023.06		50	0	0	90.62	59.38	3.12
DIQL2023.06		46.88	—	—	90.62	78.12	—
QPLEX2023.06		29.55	1.14	0	88.64	62.78	0
QMIX2023.06		13.35	0	0	75	23.01	0
VDN2023.06		—	—	47.16	—	—	—