UAI 2020poster8 citations

Finite-Memory Near-Optimal Learning for Markov Decision Processes with Long-Run Average Reward

Jan Kretinsky, Fabian Michel, Lukas Michel, Guillermo Perez

Abstract

We consider learning policies online in Markov decision processes with the long-run average reward (a.k.a. mean payoff). To ensure implementability of the policies, we focus on policies with finite memory. Firstly, we show that near optimality can be achieved almost surely, using an unintuitive gadget we call forgetfulness. Secondly, we extend the approach to a setting with partial knowledge of the system topology, introducing two optimality measures and providing near-optimal algorithms also for these cases.

BibTeX
@InProceedings{pmlr-v124-kretinsky20a,
  title = 	 {Finite-Memory Near-Optimal Learning for Markov Decision Processes with Long-Run Average Reward},
  author =       {Kretinsky, Jan and Michel, Fabian and Michel, Lukas and Perez, Guillermo},
  booktitle = 	 {Proceedings of the 36th Conference on Uncertainty in Artificial Intelligence (UAI)},
  pages = 	 {1149--1158},
  year = 	 {2020},
  editor = 	 {Peters, Jonas and Sontag, David},
  volume = 	 {124},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {03--06 Aug},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v124/kretinsky20a/kretinsky20a.pdf},
  url = 	 {https://proceedings.mlr.press/v124/kretinsky20a.html},
  abstract = 	 {We consider learning policies online in Markov decision processes with the long-run average reward (a.k.a. mean payoff). To ensure implementability of the policies, we focus on policies with finite memory. Firstly, we show that near optimality can be achieved almost surely, using an unintuitive gadget we call forgetfulness. Secondly, we extend the approach to a setting with partial knowledge of the system topology, introducing two optimality measures and providing near-optimal algorithms also for these cases.}
}