Two Views on Multiple Mean-Payoff Objectives in Markov Decision
Processes

D 2011

Two Views on Multiple Mean-Payoff Objectives in Markov Decision Processes

BROŽEK, Václav, Tomáš BRÁZDIL, Krishnendu CHATTERJEE, Vojtěch FOREJT, Antonín KUČERA et. al.

Basic information

Original name

Two Views on Multiple Mean-Payoff Objectives in Markov Decision Processes

Authors

BROŽEK, Václav (203 Czech Republic, belonging to the institution), Tomáš BRÁZDIL (203 Czech Republic, belonging to the institution), Krishnendu CHATTERJEE (356 India), Vojtěch FOREJT (203 Czech Republic, belonging to the institution) and Antonín KUČERA (203 Czech Republic, guarantor, belonging to the institution)

Edition

Los Alamitos, California, Proceedings 26th Annual IEEE Symposium on Logic in Computer Science, p. 33-42, 10 pp. 2011

Publisher

IEEE

Other information

Language

English

Type of outcome

Stať ve sborníku

Field of Study

10201 Computer sciences, information science, bioinformatics

Country of publisher

United States of America

Confidentiality degree

není předmětem státního či obchodního tajemství

Publication form

printed version "print"

RIV identification code

RIV/00216224:14330/11:00051963

Organization unit

Faculty of Informatics

ISBN

978-0-7695-4412-0

DOI

http://dx.doi.org/10.1109/LICS.2011.10

UT WoS

000297350400006

Keywords in English

Markov decision process; optimization with multiple objectives; mean payoff; Pareto curve; approximation

Abstract

ORIG CZ

V originále

We study Markov decision processes (MDPs) with multiple limit-average (or mean-payoff) functions. We consider two different objectives, namely, expectation and satisfaction objectives. Given an MDP with k reward functions, in the expectation objective the goal is to maximize the expected value, and in the satisfaction objective the goal is to maximize the probability of runs such that the limit-average value stays above a given vector.

In Czech

V článku jsou studovány Markovovy rozhodovací procesy s několika limit-average funkcemi. Uvažují se dva typy výherních podmínek a pro obě jsou podány základní výsledky o rozhodnutelnosti a složitosti příslušných rozhodovacích problémů.

Links

P202/10/1469, interní kód MU

Name: Formální metody pro analýzu a verifikaci komplexních systémů

1M0545, research and development project

Name: Institut Teoretické Informatiky

Investor: Ministry of Education, Youth and Sports of the CR, Institute for Theoretical Computer Science

Citovat

BROŽEK, Václav, Tomáš BRÁZDIL, Krishnendu CHATTERJEE, Vojtěch FOREJT and Antonín KUČERA. Two Views on Multiple Mean-Payoff Objectives in Markov Decision Processes. In Martin Grohe. Proceedings 26th Annual IEEE Symposium on Logic in Computer Science. Los Alamitos, California: IEEE, 2011, p. 33-42. ISBN 978-0-7695-4412-0. Available from: https://dx.doi.org/10.1109/LICS.2011.10.

@inproceedings{929376,
   author = {Brožek, Václav and Brázdil, Tomáš and Chatterjee, Krishnendu and Forejt, Vojtěch and Kučera, Antonín},
   address = {Los Alamitos, California},
   booktitle = {Proceedings 26th Annual IEEE Symposium on Logic in Computer Science},
   doi = {http://dx.doi.org/10.1109/LICS.2011.10},
   editor = {Martin Grohe},
   keywords = {Markov decision process; optimization with multiple objectives; mean payoff; Pareto curve; approximation},
   howpublished = {tištěná verze "print"},
   language = {eng},
   location = {Los Alamitos, California},
   isbn = {978-0-7695-4412-0},
   pages = {33-42},
   publisher = {IEEE},
   title = {Two Views on Multiple Mean-Payoff Objectives in Markov Decision Processes},
   year = {2011}
}

TY  - JOUR
ID  - 929376
AU  - Brožek, Václav - Brázdil, Tomáš - Chatterjee, Krishnendu - Forejt, Vojtěch - Kučera, Antonín
PY  - 2011
TI  - Two Views on Multiple Mean-Payoff Objectives in Markov Decision Processes
PB  - IEEE
CY  - Los Alamitos, California
SN  - 9780769544120
KW  - Markov decision process
KW  - optimization with multiple objectives
KW  - mean payoff
KW  - Pareto curve
KW  - approximation
N2  - We study Markov decision processes (MDPs) with multiple limit-average (or mean-payoff) functions. We consider two different objectives, namely, expectation and satisfaction objectives. Given an MDP with k reward functions, in the expectation objective the goal is to maximize the expected value, and in the satisfaction objective the goal is to maximize the probability of runs such that the limit-average value stays above a given vector.
ER  -

BROŽEK, Václav, Tomáš BRÁZDIL, Krishnendu CHATTERJEE, Vojtěch FOREJT and Antonín KUČERA. Two Views on Multiple Mean-Payoff Objectives in Markov Decision Processes. In Martin Grohe. \textit{Proceedings 26th Annual IEEE Symposium on Logic in Computer Science}. Los Alamitos, California: IEEE, 2011, p.~33-42. ISBN~978-0-7695-4412-0. Available from: https://dx.doi.org/10.1109/LICS.2011.10.

Detailed Information on Publication Record