Kernel-Based Least Squares Policy Iteration for Reinforcement Learning

• Published in: IEEE transaction on neural networks and learning systems 2007-07, Vol.18 (4), p.973-992
• Main Authors: Xu, Xin, Hu, Dewen, Lu, Xicheng
• Format: Article
• Language: English
• Subjects: Adaptive control; Algorithms; Approximate dynamic programming; Artificial Intelligence; Biomimetics - methods; Computer Simulation; Convergence; Decision Support Techniques; Distance learning; Dynamical systems; Feedback; Feedback control; Iterative methods; Kernel; kernel methods; Kernels; Learning; least squares; Least squares approximation; Least squares method; Least squares methods; Least-Squares Analysis; Linear approximation; Markov Chains; Markov decision problems (MDPs); Mathematical models; Models, Theoretical; Policies; Programmable control; Reinforcement (Psychology); reinforcement learning (RL); State-space methods; Studies
• ISSN: 1045-9227; 2162-237X; 1941-0093; 2162-2388
• DOI: 10.1109/TNN.2007.899161

In this paper, we present a kernel-based least squares policy iteration (KLSPI) algorithm for reinforcement learning (RL) in large or continuous state spaces, which can be used to realize adaptive feedback control of uncertain dynamic systems. By using KLSPI, near-optimal control policies can be obtained without much a priori knowledge on dynamic models of control plants. In KLSPI, Mercer kernels are used in the policy evaluation of a policy iteration process, where a new kernel-based least squares temporal-difference algorithm called KLSTD-Q is proposed for efficient policy evaluation. To keep the sparsity and improve the generalization ability of KLSTD-Q solutions, a kernel sparsification procedure based on approximate linear dependency (ALD) is performed. Compared to the previous works on approximate RL methods, KLSPI makes two progresses to eliminate the main difficulties of existing results. One is the better convergence and (near) optimality guarantee by using the KLSTD-Q algorithm for policy evaluation with high precision. The other is the automatic feature selection using the ALD-based kernel sparsification. Therefore, the KLSPI algorithm provides a general RL method with generalization performance and convergence guarantee for large-scale Markov decision problems (MDPs). Experimental results on a typical RL task for a stochastic chain problem demonstrate that KLSPI can consistently achieve better learning efficiency and policy quality than the previous least squares policy iteration (LSPI) algorithm. Furthermore, the KLSPI method was also evaluated on two nonlinear feedback control problems, including a ship heading control problem and the swing up control of a double-link underactuated pendulum called acrobot. Simulation results illustrate that the proposed method can optimize controller performance using little a priori information of uncertain dynamic systems. It is also demonstrated that KLSPI can be applied to online learning control by incorporating an initial controller to ensure online performance.