Multi-Agent Reinforcement Learning on SMAC 3s_vs_3z benchmark leaderboard · SOTA2 Research

Multi-Agent Reinforcement Learning on SMAC 3s_vs_3z

93.5Dis-1IBAL

Evaluation Results

Method	Links
IBAL2026.05		93.5	99.7	99.9	81.7	93.7
ROMANCE2026.05		55.4	98	96.8	57.1	76.1
WALL2026.05		37.5	98.8	97.8	68.8	75.7
FGSM2026.05		25	95.8	96.2	41.2	64.6
Rand-Obs2026.05		22.1	96.3	88.1	29.2	58.9
Rand-Act2026.05		12.5	90.7	91.3	30.2	56.2
ERNIE2026.05		6.2	98	92.1	45.4	60.4
Vanilla QMIX2026.05		5.2	89	76.7	2.1	43.2
ATLA2026.05		3.1	98.2	96.4	43.8	60.9