Multi-Agent Reinforcement Learning on Simple Spread N=3 benchmark leaderboard · SOTA2 Research

Multi-Agent Reinforcement Learning on Simple Spread N=3

0.0017CollisionsCG-CMARL

Evaluation Results

Method	Links
CG-CMARL2026.06		0.0017	0.0006	3.92
CG-CMARL2026.06		0.0022	0.0007	1.33
CG-CMARL2026.06		0.003	0.001	8.25
CG-CMARL2026.06		0.0034	0.0011	17.5
CG-CMARL2026.06		0.0037	0.0012	46.17
CG-CMARL2026.06		0.0042	0.0014	38.08
CG-CMARL2026.06		0.0052	0.0017	47.33
QMIX2026.06		0.0052	0.0017	7.83
CG-CMARL2026.06		0.0054	0.0018	46.58
CG-CMARL2026.06		0.0055	0.0018	47.58
QMIX2026.06		0.006	0.002	5.6
MAPPO-Lag2026.06		0.0067	0.0022	9.97
QMIX2026.06		0.0071	0.0024	5.83
MAPPO2026.06		0.0084	0.0028	10.9
MAPPO2026.06		0.0126	0.0042	8
MAPPO-Lag2026.06		0.0185	0.0062	10.53
MAPPO2026.06		0.0232	0.0077	9.84
DCG2026.06		0.0322	0.0107	24.17
DCG2026.06		0.0439	0.0146	23.5
DCG2026.06		0.045	0.015	21.58
IQL2026.06		0.0483	0.0161	15.67
IQL2026.06		0.0519	0.0173	20.58
IQL2026.06		0.0623	0.0208	17.42