Faculty of Business Administration
Visiting Scholar Seminar
Risk-Sensitive Finite-Horizon Piecewise Deterministic Markov Decision Processes and Empirical Analysis
Yonghui HUANG
School of Mathematics
Zhongshan University
Guangzhou
Date: 19/08/2021, Thursday
Time: 3:00pm – 4:30pm
Zoom:
https://umac.zoom.us/j/96544231587?pwd=UXl5UE1HVHpRTERJQkthZWg1RG1wQT09
Password: 619437
Abstract
In this talk, we discuss zero-sum piecewise deterministic Markov games with Borel state and action spaces, where the expected infinite-horizon discounted payoff criterion is considered. Both the transition rate and payoff function could be unbounded. The policies of both players are history-dependent, and the controls continuously act on the transition rate and the payoff rate. Under suitable conditions, the non-explosion of the process, we develop the finiteness of the criterion and the Dynkin’s formula in our setup, via which we show that the value function of the zero-sum game is the unique solution to the Shapley equation with the form of a differential equation. Meanwhile, the existence of Nash equilibrium is established.
Biography
Professor Yonghui Huang is an Associate Professor in the School of Mathematics, Zhongshan University, Guangzhou. He received his PhD degree in Probability and Statistics from the same school in Zhongshan University. Prof. Huang’s research interests are Markovian decision process and its applications. Prof. Huang has published 20 research papers in outstanding journals, including Mathematics of Operations Research, Journal of Applied Probability, Advanced Applied Probability.
