DMS Applied Mathematics Seminar

PLEASE NOTE NEW TIME AND NEW PLACE

Speaker: Dr. Bo Liu, Computer Science Department, AU

Title: Gradient, Semi-gradient and Pseudo-gradient Reinforcement Learning

Abstract: In this talk, I will present the establishment of a unified general framework for stochastic-gradient-based temporal-difference learning algorithms that use proximal gradient methods. The primal-dual saddle-point formulation is introduced, and state-of-the-art stochastic gradient solvers, such as mirror descent and extragradient are used to design several novel reinforcement learning algorithms. The finite-sample analysis is given along with detailed empirical experiments to demonstrate the effectiveness of the proposed algorithms. Several important extensions, such as control learning, variance reduction, acceleration, and regularization will also be discussed in detail.