EAQR: A Multiagent Q-Learning Algorithm for Coordination of Multiple Agents

Table 4

Success rate for the DSN problem (evaluation episodes = 5000).

	= 10,000	= 50,000	= 100,000
EAQR	47.4%	99.9%	100%
WoLF-PHC	22.8%	34.1%	33.7%
EMA Q-learning	7.8%	6.8%	7.1%
Single-agent RL	0	0	0