Share your thoughts, 1 month free Claude Pro on usSee more

Cooperative Multi-Agent Reinforcement Learning

Benchmarks

Dataset Name	SOTA Method	Metric
Heterogeneous Cloud Scheduling 40 scenarios (test)	DG-PG	Best Checkpoint Reward-40.5	18	3mo ago
Disperse (last 2% of train)	DICG	Mean Episodic Reward-0.36	13	22d ago
Adversary (last 2% of train)	SACHI	Mean Episodic Reward85.04	13	22d ago
Speaker-Listener (last 2% of train)	SACHI	Mean Episodic Reward-17.22	13	22d ago
Crypto (last 2% of train)	DCG	Mean Episodic Reward50	13	22d ago
Reference (last 2% of train)	SACHI	Mean Episodic Reward-25.39	13	22d ago
SustainGym BuildingEnv climatic and seasonal shifts	robust QTRAN	Normalized Episodic Returns73.3	12	3mo ago
SustainGym BuildingEnv season_2 (test)	robust QMIX	Normalized Episodic Return91.6	12	3mo ago
SMAC 3s5z map	QMIX	Return18.7	11	1mo ago
SMAC 5m_vs_6m map	QMIX	Return19.1	11	1mo ago
SMAC 8m map	KD-MARL	Return17.8	11	1mo ago
SMAC 3m map		Return19.8	11	1mo ago
SMAC 3s5z_vs_3s6z (SH)	AlberDICE	Mean Success Rate63	6	3mo ago
SMAC 8m_vs_9m Hard	AlberDICE	Mean Success Rate0.67	6	3mo ago
SMAC 6hvs8z SH	AlberDICE	Mean Success Rate21	6	3mo ago
SMAC Corridor (SH)	AlberDICE	Mean Success Rate98	6	3mo ago
SMAC 5m_vs_6m Hard	AlberDICE	Mean Success Rate0.24	6	3mo ago
SMAC 3s5z Hard	AlberDICE	Mean Success Rate47	6	3mo ago
Overcooked-AI Forced Coordination (test)	IAD	Mean Reward35.43	4	8d ago
Overcooked-AI Counter Circuit (test)	IAD	Mean Reward43.65	4	8d ago
Overcooked-AI Coordination Ring (test)	IAD	Mean Reward97.35	4	8d ago
Overcooked-AI Asymmetric Advantages (test)	IAD	Mean Reward124.39	4	8d ago
Overcooked-AI Cramped Room (test)	IAD	Mean Reward148.76	4	8d ago
Rover-Tower 4 rovers and 4 towers scenario	TransfQmix	FComm40	4	14d ago
SMAX	MARS	IQM of Normalized Returns0.92	3	22d ago

Showing 25 of 34 rows