Controlled Markov Decision Processes with AVaR criteria for unbounded costs

https://doi.org/10.1016/j.cam.2016.11.052Get rights and content
Under an Elsevier user license
open archive

Abstract

In this paper, we consider the control problem with the Average-Value-at-Risk (AVaR) criteria of the possibly unbounded L1-costs in infinite horizon on a Markov Decision Process (MDP). With a suitable state aggregation and by choosing a priori a global variable s heuristically, we show that there exist optimal policies for the infinite horizon problem for possibly unbounded costs.

MSC

90C39
93E20

Keywords

Markov decision problem
Average-Value-at-Risk
Optimal control

Cited by (0)