一种基于模型的可分解贝叶斯在线强化学习

针对贝叶斯强化学习中参数个数巨大，收敛速度慢，无法实现在线学习的问题，提出一种基于模型的可分解贝叶斯强化学习方法.首先，将学习参数进行可分解表示，降低学习参数的个数；然后，根据先验知识和观察数据采用贝叶斯方法来学习，最优化探索和利用二者之间的平衡关系；最后，采用基于点的贝叶斯强化学习方法实现学习过程的快速收敛，从而达到在线学习的目的.仿真结果表明该算法能够满足实时系统性能的要求.

Model-Based Factored Bayesian Online Reinforcement Learning

Due to the enormous number of parameters and slow convergence which are the major obstacles for online learning in model-based Bayesian reinforcement learning,the paper presents a model-based factored Bayesian reinforcement learning approach.Firstly,factored representations are made to represent the dynamics with fewer parameters.Then,according to prior knowledge and observable data,this paper exploits model-based reinforcement learning to provide an elegant solution to the optimal exploration-exploitation tradeoff.Finally,a pointed-based Bayesian reinforcement learning approach is proposed to speed up the convergence to achieve online learning.The experimental results show that the proposed approach can approximate the underlying Bayesian reinforcement learning task well with guaranteed real-time performance.