Skip to main navigation Skip to search Skip to main content

A unified approach for semi-Markov decision processes with discounted and average reward criteria

  • Harbin Institute of Technology Shenzhen
  • Shenzhen Institute of Information Technology

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

Abstract

On the basis of the sensitivity-based optimization, we develop a unified optimization approach for semi-Markov decision processes (SMDPs) with infinite horizon discounted and average reward criteria. We show that the sensitivity formula under average reward criteria is a limitation case of discounted reward criteria. On the basis of the performance sensitivity formulas, we provide a unified formulation for the policy iteration algorithms of semi-Markov decision processes with discounted and average reward criteria.

Original languageEnglish
Title of host publicationProceeding of the 11th World Congress on Intelligent Control and Automation, WCICA 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1741-1744
Number of pages4
EditionMarch
ISBN (Electronic)9781479958252
DOIs
StatePublished - 2 Mar 2015
Externally publishedYes
Event2014 11th World Congress on Intelligent Control and Automation, WCICA 2014 - Shenyang, China
Duration: 29 Jun 20144 Jul 2014

Publication series

NameProceedings of the World Congress on Intelligent Control and Automation (WCICA)
NumberMarch
Volume2015-March

Conference

Conference2014 11th World Congress on Intelligent Control and Automation, WCICA 2014
Country/TerritoryChina
CityShenyang
Period29/06/144/07/14

Keywords

  • Performance difference
  • Policy iteration
  • SMDPs

Fingerprint

Dive into the research topics of 'A unified approach for semi-Markov decision processes with discounted and average reward criteria'. Together they form a unique fingerprint.

Cite this