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

Table 7

Average steps for the DSN problem (evaluation episodes = 5000).

	= 10,000	= 50,000	= 100,000
EAQR	3.65 ± 0.35	3.22 ± 0.26	3.12 ± 0.20
WoLF-PHC	3.64 ± 0.63	3.58 ± 0.64	3.69 ± 0.61
EMA Q-learning	3.81 ± 0.43	3.94 ± 0.44	3.93 ± 0.45
Single-agent RL	5.57 ± 0.25	5.3 ± 0.29	5.06 ± 0.27