Restless Bandits with Average Reward: Breaking the Uniform Global Attractor Assumption

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hong, Yige, Xie, Qiaomin, Chen, Yudong, Wang, Weina
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913195089723392
author Hong, Yige
Xie, Qiaomin
Chen, Yudong
Wang, Weina
author_facet Hong, Yige
Xie, Qiaomin
Chen, Yudong
Wang, Weina
contents We study the infinite-horizon restless bandit problem with the average reward criterion, in both discrete-time and continuous-time settings. A fundamental goal is to efficiently compute policies that achieve a diminishing optimality gap as the number of arms, $N$, grows large. Existing results on asymptotic optimality all rely on the uniform global attractor property (UGAP), a complex and challenging-to-verify assumption. In this paper, we propose a general, simulation-based framework, Follow-the-Virtual-Advice, that converts any single-armed policy into a policy for the original $N$-armed problem. This is done by simulating the single-armed policy on each arm and carefully steering the real state towards the simulated state. Our framework can be instantiated to produce a policy with an $O(1/\sqrt{N})$ optimality gap. In the discrete-time setting, our result holds under a simpler synchronization assumption, which covers some problem instances that violate UGAP. More notably, in the continuous-time setting, we do not require \emph{any} additional assumptions beyond the standard unichain condition. In both settings, our work is the first asymptotic optimality result that does not require UGAP.
format Preprint
id arxiv_https___arxiv_org_abs_2306_00196
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Restless Bandits with Average Reward: Breaking the Uniform Global Attractor Assumption
Hong, Yige
Xie, Qiaomin
Chen, Yudong
Wang, Weina
Machine Learning
Optimization and Control
Probability
90C40
G.3; I.6
We study the infinite-horizon restless bandit problem with the average reward criterion, in both discrete-time and continuous-time settings. A fundamental goal is to efficiently compute policies that achieve a diminishing optimality gap as the number of arms, $N$, grows large. Existing results on asymptotic optimality all rely on the uniform global attractor property (UGAP), a complex and challenging-to-verify assumption. In this paper, we propose a general, simulation-based framework, Follow-the-Virtual-Advice, that converts any single-armed policy into a policy for the original $N$-armed problem. This is done by simulating the single-armed policy on each arm and carefully steering the real state towards the simulated state. Our framework can be instantiated to produce a policy with an $O(1/\sqrt{N})$ optimality gap. In the discrete-time setting, our result holds under a simpler synchronization assumption, which covers some problem instances that violate UGAP. More notably, in the continuous-time setting, we do not require \emph{any} additional assumptions beyond the standard unichain condition. In both settings, our work is the first asymptotic optimality result that does not require UGAP.
title Restless Bandits with Average Reward: Breaking the Uniform Global Attractor Assumption
topic Machine Learning
Optimization and Control
Probability
90C40
G.3; I.6
url https://arxiv.org/abs/2306.00196