进化博弈中多代理人强化学习模型

Multiagent reinforcement learning-model in evolutionary games

将强化学习引入到进化博弈中,建立了进化博弈中的多代理人强化学习模型,并基于Q-学习给出了算法流程,仿真算例的结果表明多代理人强化学习模型能使得博弈人不断学习、寻求最优策略.

In this paper,reinforcement learning is introduced in evolutionary games,multiagent reinforcement learning-model is presented,and the learning algorithm is given based on Q-learning.The results of simulation experiments show that the multiagent reinforcement learning-model can make agents find the optimal strategy by learning.