Dynamic analysis of multiagent Qlearning withεgreedy exploration DynamicanalysisofmultiagentQ-learningwithε-greedyexploration