Speaker：Qingda Wei (Huaqiao University)

Time：2023-5-26, 16:30

Location：Conference Room 109 at Experiment Building at Haiyun Campus

Abstract:

In this talk, we will discuss some recent progress on discrete-time Markov decision processes under risk-sensitive average cost criterion. The state space is a Borel space and the cost function can be possibly unbounded. Under suitable conditions we derive the multiplicative Poisson equation. Moreover, we establish the existence of a solution to risk-sensitive average cost optimality equation and the existence of a stationary optimal policy. Finally, we give a policy iteration algorithm and analyze its convergence.