Controlled Markov Chains with AVaR Criteria for Unbounded Costs

Autor:	Ugurlu, Kerem
Rok vydání:	2015
Předmět:	Mathematics - Probability Mathematics - Optimization and Control
Druh dokumentu:	Working Paper
Popis:	In this paper, we consider the control problem with the Average-Value-at-Risk (AVaR) criteria of the possibly unbounded $L^{1}$-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. To our knowledge, this is the first work of deriving dynamic programming equations with $L^1$-unbounded costs via AVaR-operator.
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/1501.02518 Zobrazit plný text záznamu View this record from Arxiv