题 目：Mean-variance problems for discounted Markov decision processes
报 告 人：郭先平 教授
In this talk, we introduce so-called mean-variance problems for Markov decision processes (MDPs) over the set of deterministic stationary policies. Different from the usual formulation in MDPs, we aim to obtain a so-called mean-variance optimal policy that minimizes the variance over a set of all policies with a given mean (expected) reward. For the discounted continuous-time MDPs with finite-state and action spaces, we will show the existence and calculation of a mean-variance optimal policy, and also talk about some further topics.