Skip to main navigation Skip to search Skip to main content

Markov decision processes and stochastic games with total effective payoff

  • Rutgers University
  • Kyoto University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

6 Scopus citations

Abstract

We consider finite Markov decision processes (MDPs) with undiscounted total effective payoff. We show that there exist uniformly optimal pure stationary strategies that can be computed by solving a polynomial number of linear programs. We apply this result to two-player zero-sum stochastic games with perfect information and undiscounted total effective payoff, and derive the existence of a saddle point in uniformly optimal pure stationary strategies.

Original languageBritish English
Title of host publication32nd International Symposium on Theoretical Aspects of Computer Science, STACS 2015
EditorsErnst W. Mayr, Nicolas Ollinger
Pages103-115
Number of pages13
ISBN (Electronic)9783939897781
DOIs
StatePublished - 1 Feb 2015
Event32nd International Symposium on Theoretical Aspects of Computer Science, STACS 2015 - Garching, Germany
Duration: 4 Mar 20157 Mar 2015

Publication series

NameLeibniz International Proceedings in Informatics, LIPIcs
Volume30
ISSN (Print)1868-8969

Conference

Conference32nd International Symposium on Theoretical Aspects of Computer Science, STACS 2015
Country/TerritoryGermany
CityGarching
Period4/03/157/03/15

Keywords

  • Linear programming
  • Markov decision processes
  • Mean payoff
  • Total payoff
  • Undiscounted stochastic games

Fingerprint

Dive into the research topics of 'Markov decision processes and stochastic games with total effective payoff'. Together they form a unique fingerprint.

Cite this