基于Q-学习的进化博弈决策模型

基于Q-强化学习算法,建立了进化博弈中代理人的决策模型.考虑到强化学习算法不需要建立环境模型,可用于不完全、不确定信息问题,将Q-强化学习算法引入到进化博弈中,研究了进化博弈中两种Q-学习决策模型:单代理人Q-学习决策模型和多代理人Q-学习决策模型,并针对不同结构的进化博弈选择不同的决策模型和算法进行了讨论.仿真算例的结果说明基于Q-学习的决策模型能指导代理人学习、选择最优策略.

Decision-models in evolutionary games based on Q-learning

Based on Q-reinforcement learning,decision-models of agents in evolutionary games are established.Considering that the reinforcement learning does not need a model of its environment,and it can be used in problems with incomplete and uncertain information,Q-learning is introduced to evolutionary games;and the single-agent Q-learning decision-model and the multi-agent Q-learning decision-model are presented.In addition,how to choose decision-models and algorithms according to the type of games is discussed.The results of simulation experiments show that the decision-models based on Q-learning can make agents to choose the optimal strategy by learning.