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

Table 6

Minimal cumulative reward for the DSN problem (evaluation episodes = 5000).

	= 10,000	= 50,000	= 100,000
EAQR	39.26	41.97	42
WoLF-PHC	37.65	38.71	38.82
EMA Q-learning	32.92	32.08	32.53
Single-agent RL	25.12	29.38	32.21