Average Cost Optimality of Partially Observed MDPS: Contraction of Non-linear Filters, Optimal Solutions and Approximations

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Demirci, Yunus Emre, Kara, Ali Devran, Yüksel, Serdar
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914893392773120
author Demirci, Yunus Emre
Kara, Ali Devran
Yüksel, Serdar
author_facet Demirci, Yunus Emre
Kara, Ali Devran
Yüksel, Serdar
contents The average cost optimality is known to be a challenging problem for partially observable stochastic control, with few results available beyond the finite state, action, and measurement setup, for which somewhat restrictive conditions are available. In this paper, we present explicit and easily testable conditions for the existence of solutions to the average cost optimality equation where the state space is compact. In particular, we present a new contraction based analysis, which is new to the literature to our knowledge, building on recent regularity results for non-linear filters. Beyond establishing existence, we also present several implications of our analysis that are new to the literature: (i) robustness to incorrect priors (ii) near optimality of policies based on quantized approximations, (iii) near optimality of policies with finite memory, and (iv) convergence in Q-learning. In addition to our main theorem, each of these represents a novel contribution for average cost criteria.
format Preprint
id arxiv_https___arxiv_org_abs_2312_14111
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Average Cost Optimality of Partially Observed MDPS: Contraction of Non-linear Filters, Optimal Solutions and Approximations
Demirci, Yunus Emre
Kara, Ali Devran
Yüksel, Serdar
Optimization and Control
90C40, 93E11, 93E20
The average cost optimality is known to be a challenging problem for partially observable stochastic control, with few results available beyond the finite state, action, and measurement setup, for which somewhat restrictive conditions are available. In this paper, we present explicit and easily testable conditions for the existence of solutions to the average cost optimality equation where the state space is compact. In particular, we present a new contraction based analysis, which is new to the literature to our knowledge, building on recent regularity results for non-linear filters. Beyond establishing existence, we also present several implications of our analysis that are new to the literature: (i) robustness to incorrect priors (ii) near optimality of policies based on quantized approximations, (iii) near optimality of policies with finite memory, and (iv) convergence in Q-learning. In addition to our main theorem, each of these represents a novel contribution for average cost criteria.
title Average Cost Optimality of Partially Observed MDPS: Contraction of Non-linear Filters, Optimal Solutions and Approximations
topic Optimization and Control
90C40, 93E11, 93E20
url https://arxiv.org/abs/2312.14111