Optimizing Local Satisfaction of Long-Run Average Objectives in
Markov Decision Processes

D 2024

Optimizing Local Satisfaction of Long-Run Average Objectives in Markov Decision Processes

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

Základní údaje

Originální název

Optimizing Local Satisfaction of Long-Run Average Objectives in Markov Decision Processes

Autoři

KLAŠKA, David (203 Česká republika, domácí), Antonín KUČERA (203 Česká republika, garant, domácí), Vojtěch KŮR (203 Česká republika, domácí), Vít MUSIL (203 Česká republika, domácí) a Vojtěch ŘEHÁK (203 Česká republika, domácí)

Vydání

Washington, DC,. Neuveden, Proceedings of 38th Annual AAAI Conference on Artificial Intelligence (AAAI 2024), od s. 20143-20150, 8 s. 2024

Nakladatel

AAAI Press

Další údaje

Jazyk

angličtina

Typ výsledku

Stať ve sborníku

Obor

10201 Computer sciences, information science, bioinformatics

Utajení

není předmětem státního či obchodního tajemství

Forma vydání

tištěná verze "print"

Odkazy

Paper URL

Organizační jednotka

Fakulta informatiky

ISBN

978-1-57735-887-9

ISSN

DOI

http://dx.doi.org/10.1609/aaai.v38i18.29993

Klíčová slova anglicky

Markov decision processes; invariant distribution

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 26. 7. 2024 11:27, RNDr. Pavel Šmerk, Ph.D.

Anotace

V originále

Long-run average optimization problems for Markov decision processes (MDPs) require constructing policies with optimal steady-state behavior, i.e., optimal limit frequency of visits to the states. However, such policies may suffer from local instability in the sense that the frequency of states visited in a bounded time horizon along a run differs significantly from the limit frequency. In this work, we propose an efficient algorithmic solution to this problem.

Návaznosti

0011629866, interní kód MU

Název: Models, Algorithms, and Tools for Solving Adversarial Security Problems

Investor: Ostatní - zahraniční, Models, Algorithms, and Tools for Solving Adversarial Security Problems

101087529, interní kód MU

Název: Cyber-security Excellence Hub in Estonia and South Moravia (CHESS)

Investor: Evropská unie, Cyber-security Excellence Hub in Estonia and South Moravia (CHESS), Rozšiřování účasti a posílení ERA

Přiložené soubory

paper.pdf

Vyhledat podobné dokumenty

Citovat

KLAŠKA, David, Antonín KUČERA, Vojtěch KŮR, Vít MUSIL a Vojtěch ŘEHÁK. Optimizing Local Satisfaction of Long-Run Average Objectives in Markov Decision Processes. In Wooldridge M., Dy J., Natarajan S. Proceedings of 38th Annual AAAI Conference on Artificial Intelligence (AAAI 2024). Washington, DC,. Neuveden: AAAI Press, 2024, s. 20143-20150. ISBN 978-1-57735-887-9. Dostupné z: https://dx.doi.org/10.1609/aaai.v38i18.29993.

@inproceedings{2396817,
   author = {Klaška, David and Kučera, Antonín and Kůr, Vojtěch and Musil, Vít and Řehák, Vojtěch},
   address = {Neuveden},
   booktitle = {Proceedings of 38th Annual AAAI Conference on Artificial Intelligence (AAAI 2024)},
   doi = {http://dx.doi.org/10.1609/aaai.v38i18.29993},
   edition = {Washington, DC,},
   editor = {Wooldridge M., Dy J., Natarajan S.},
   keywords = {Markov decision processes; invariant distribution},
   howpublished = {tištěná verze "print"},
   language = {eng},
   location = {Neuveden},
   isbn = {978-1-57735-887-9},
   pages = {20143-20150},
   publisher = {AAAI Press},
   title = {Optimizing Local Satisfaction of Long-Run Average Objectives in Markov Decision Processes},
   url = {https://ojs.aaai.org/index.php/AAAI/article/view/29993},
   year = {2024}
}

TY  - JOUR
ID  - 2396817
AU  - Klaška, David - Kučera, Antonín - Kůr, Vojtěch - Musil, Vít - Řehák, Vojtěch
PY  - 2024
TI  - Optimizing Local Satisfaction of Long-Run Average Objectives in Markov Decision Processes
PB  - AAAI Press
CY  - Neuveden
SN  - 9781577358879
KW  - Markov decision processes
KW  - invariant distribution
UR  - https://ojs.aaai.org/index.php/AAAI/article/view/29993
N2  - Long-run average optimization problems for Markov decision processes (MDPs) require constructing policies with optimal steady-state behavior, i.e., optimal limit frequency of visits to the states. However, such policies may suffer from local instability in the sense that the frequency of states visited in a bounded time horizon along a run differs significantly from the limit frequency. In this work, we propose an efficient algorithmic solution to this problem.
ER  -

KLAŠKA, David, Antonín KUČERA, Vojtěch KŮR, Vít MUSIL a Vojtěch ŘEHÁK. Optimizing Local Satisfaction of Long-Run Average Objectives in Markov Decision Processes. In Wooldridge M., Dy J., Natarajan S. \textit{Proceedings of 38th Annual AAAI Conference on Artificial Intelligence (AAAI 2024)}. Washington, DC,. Neuveden: AAAI Press, 2024, s.~20143-20150. ISBN~978-1-57735-887-9. Dostupné z: https://dx.doi.org/10.1609/aaai.v38i18.29993.

Podrobný výpis o publikaci