Optimal Non-Asymptotic Rates of Value Iteration for Average-Reward Markov Decision Processes

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lee, Jongmin, Ryu, Ernest K.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908821211840512
author Lee, Jongmin
Ryu, Ernest K.
author_facet Lee, Jongmin
Ryu, Ernest K.
contents While there is an extensive body of research on the analysis of Value Iteration (VI) for discounted cumulative-reward MDPs, prior work on analyzing VI for (undiscounted) average-reward MDPs has been limited, and most prior results focus on asymptotic rates in terms of Bellman error. In this work, we conduct refined non-asymptotic analyses of average-reward MDPs, obtaining a collection of convergence results that advance our understanding of the setup. Among our new results, most notable are the $\mathcal{O}(1/k)$-rates of Anchored Value Iteration on the Bellman error under the multichain setup and the span-based complexity lower bound that matches the $\mathcal{O}(1/k)$ upper bound up to a constant factor of $8$ in the weakly communicating and unichain setups
format Preprint
id arxiv_https___arxiv_org_abs_2504_09913
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Optimal Non-Asymptotic Rates of Value Iteration for Average-Reward Markov Decision Processes
Lee, Jongmin
Ryu, Ernest K.
Optimization and Control
While there is an extensive body of research on the analysis of Value Iteration (VI) for discounted cumulative-reward MDPs, prior work on analyzing VI for (undiscounted) average-reward MDPs has been limited, and most prior results focus on asymptotic rates in terms of Bellman error. In this work, we conduct refined non-asymptotic analyses of average-reward MDPs, obtaining a collection of convergence results that advance our understanding of the setup. Among our new results, most notable are the $\mathcal{O}(1/k)$-rates of Anchored Value Iteration on the Bellman error under the multichain setup and the span-based complexity lower bound that matches the $\mathcal{O}(1/k)$ upper bound up to a constant factor of $8$ in the weakly communicating and unichain setups
title Optimal Non-Asymptotic Rates of Value Iteration for Average-Reward Markov Decision Processes
topic Optimization and Control
url https://arxiv.org/abs/2504.09913