ML Anthology

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

David Klaska, Antonín Kucera, Vojtech Kur, Vít Musil, Vojtech Rehák
AAAI 2024 pp. 20143-20150
doi:10.1609/AAAI.V38I18.29993 /aaai/2024/klaska2024aaai-optimizing/

Abstract

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.

PDF AAAI Semantic Scholar

Cite

Text

Klaska et al. "Optimizing Local Satisfaction of Long-Run Average Objectives in Markov Decision Processes." AAAI Conference on Artificial Intelligence, 2024. doi:10.1609/AAAI.V38I18.29993

Markdown

[Klaska et al. "Optimizing Local Satisfaction of Long-Run Average Objectives in Markov Decision Processes." AAAI Conference on Artificial Intelligence, 2024.](https://mlanthology.org/aaai/2024/klaska2024aaai-optimizing/) doi:10.1609/AAAI.V38I18.29993

BibTeX

@inproceedings{klaska2024aaai-optimizing,
  title     = {{Optimizing Local Satisfaction of Long-Run Average Objectives in Markov Decision Processes}},
  author    = {Klaska, David and Kucera, Antonín and Kur, Vojtech and Musil, Vít and Rehák, Vojtech},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2024},
  pages     = {20143-20150},
  doi       = {10.1609/AAAI.V38I18.29993},
  url       = {https://mlanthology.org/aaai/2024/klaska2024aaai-optimizing/}
}