ResearchBenchmarksSampling on VMASFollow34.86Mean Episodic RewardCoHetteam17.12821.731526.33530.9385Aug 12, 2024Evaluation ResultsMethodMethodLinksMean Episodic RewardCoHetteamEnvironment Steps=2×10^5Environment Steps=2×10^52024.0834.86CoHetselfEnvironment Steps=2×10^5Environment Steps=2×10^52024.0831.75IPPOEnvironment Steps=2×10^5Environment Steps=2×10^52024.0826.13HetGPPOEnvironment Steps=2×10^5Environment Steps=2×10^52024.0817.81