Multiple Mean-Payoff Optimization Under Local Stability
Constraints

D 2025

Multiple Mean-Payoff Optimization Under Local Stability Constraints

KLAŠKA, David; Antonín KUČERA; Vojtěch KŮR; Vít MUSIL; Vojtěch ŘEHÁK et al.

Základní údaje

Originální název

Multiple Mean-Payoff Optimization Under Local Stability Constraints

Autoři

KLAŠKA, David; Antonín KUČERA; Vojtěch KŮR ; Vít MUSIL a Vojtěch ŘEHÁK

Vydání

Palo Alto, Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 39 No. 25: AAAI-25 Technical Tracks 25, od s. 26551-26558, 8 s. 2025

Nakladatel

AAAI

Další údaje

Jazyk

angličtina

Typ výsledku

Stať ve sborníku

Obor

10201 Computer sciences, information science, bioinformatics

Stát vydavatele

Spojené státy

Utajení

není předmětem státního či obchodního tajemství

Forma vydání

elektronická verze "online"

Odkazy

DOI

Označené pro přenos do RIV

Ano

Kód RIV

RIV/00216224:14330/25:00141698

Organizační jednotka

Fakulta informatiky

ISBN

978-1-57735-897-8

ISSN

Klíčová slova anglicky

Multiple Mean-Payoff Optimization

Štítky

core_A, firank_1, formela-sec

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 18. 3. 2026 19:08, prof. RNDr. Antonín Kučera, Ph.D.

Anotace

V originále

The long-run average payoff per transition (mean payoff) is the main tool for specifying the performance and dependability properties of discrete systems. The problem of constructing a controller (strategy) simultaneously optimizing several mean payoffs has been deeply studied for stochastic and game-theoretic models. One common issue of the constructed controllers is the instability of the mean payoffs, measured by the deviations of the average rewards per transition computed in a finite "window" sliding along a run. Unfortunately, the problem of simultaneously optimizing the mean payoffs under local stability constraints is computationally hard, and the existing works do not provide a practically usable algorithm even for non-stochastic models such as two-player games. In this paper, we design and evaluate the first efficient and scalable solution to this problem applicable to Markov decision processes.

Návaznosti

MUNI/A/1666/2024, interní kód MU

Název: Zapojení studentů Fakulty informatiky do mezinárodní vědecké komunity 25

Investor: Masarykova univerzita, Zapojení studentů Fakulty informatiky do mezinárodní vědecké komunity 25

Citovat

KLAŠKA, David; Antonín KUČERA; Vojtěch KŮR; Vít MUSIL a Vojtěch ŘEHÁK. Multiple Mean-Payoff Optimization Under Local Stability Constraints. Online. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 39 No. 25: AAAI-25 Technical Tracks 25. Palo Alto: AAAI, 2025, s. 26551-26558. ISBN 978-1-57735-897-8. Dostupné z: https://doi.org/10.1609/aaai.v39i25.34856.

@inproceedings{2510597,
   author = {Klaška, David and Kučera, Antonín and Kůr, Vojtěch and Musil, Vít and Řehák, Vojtěch},
   address = {Palo Alto},
   booktitle = {Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 39 No. 25: AAAI-25 Technical Tracks 25},
   doi = {https://doi.org/10.1609/aaai.v39i25.34856},
   keywords = {Multiple Mean-Payoff Optimization},
   howpublished = {elektronická verze "online"},
   language = {eng},
   location = {Palo Alto},
   isbn = {978-1-57735-897-8},
   pages = {26551-26558},
   publisher = {AAAI},
   title = {Multiple Mean-Payoff Optimization Under Local Stability Constraints},
   url = {http://dx.doi.org/10.1609/aaai.v39i25.34856},
   year = {2025}
}

TY  - CONF
ID  - 2510597
AU  - Klaška, David - Kučera, Antonín - Kůr, Vojtěch - Musil, Vít - Řehák, Vojtěch
PY  - 2025
TI  - Multiple Mean-Payoff Optimization Under Local Stability Constraints
PB  - AAAI
CY  - Palo Alto
SN  - 9781577358978
KW  - Multiple Mean-Payoff Optimization
UR  - http://dx.doi.org/10.1609/aaai.v39i25.34856
N2  - The long-run average payoff per transition (mean payoff) is the main tool for specifying the performance and dependability properties of discrete systems. The problem of constructing a controller (strategy) simultaneously optimizing several mean payoffs has been deeply studied for stochastic and game-theoretic models. One common issue of the constructed controllers is the instability of the mean payoffs, measured by the deviations of the average rewards per transition computed in a finite "window" sliding along a run. Unfortunately, the problem of simultaneously optimizing the mean payoffs under local stability constraints is computationally hard, and the existing works do not provide a practically usable algorithm even for non-stochastic models such as two-player games. In this paper, we design and evaluate the first efficient and scalable solution to this problem applicable to Markov decision processes.
ER  -

KLAŠKA, David; Antonín KUČERA; Vojtěch KŮR; Vít MUSIL a Vojtěch ŘEHÁK. Multiple Mean-Payoff Optimization Under Local Stability Constraints. Online. In \textit{Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 39 No. 25: AAAI-25 Technical Tracks 25}. Palo Alto: AAAI, 2025, s.~26551-26558. ISBN~978-1-57735-897-8. Dostupné z: https://doi.org/10.1609/aaai.v39i25.34856.

Přehled o publikaci