An online prediction algorithm for reinforcement learning with linear function approximation using cross entropy method

作者：Joseph Ajin George^*; Bhatnagar Shalabh

来源：Machine Learning, 2018, 107(8-10): 1385-1429.

DOI：10.1007/s10994-018-5727-z

• Markov decision process
• Prediction problem
• Reinforcement learning
• Stochastic approximation algorithm
• Cross entropy method
• Linear function approximation
• ODE method

In this paper, we provide two new stable online algorithms for the problem of prediction in reinforcement learning, i.e., estimating the value function of a model-free Markov reward process using the linear function approximation architecture and with memory and computation costs scaling quadratically in the size of the feature set. The algorithms employ the multi-timescale stochastic approximation variant of the very popular cross entropy optimization method which is a model based search method to find the global optimum of a real-valued function. A proof of convergence of the algorithms using the ODE method is provided. We supplement our theoretical results with experimental comparisons. The algorithms achieve good performance fairly consistently on many RL benchmark problems with regards to computational efficiency, accuracy and stability.

• 出版日期2018-9