Saved in:
Bibliographic Details
Main Author: Wang, Shengbo
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.03523
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908865175486464
author Wang, Shengbo
author_facet Wang, Shengbo
contents We study reinforcement learning in infinite-horizon discounted Markov decision processes with continuous state spaces, where data are generated online from a single trajectory under a Markovian behavior policy. To avoid maintaining an infinite-dimensional, function-valued estimate, we propose the novel Q-Measure-Learning, which learns a signed empirical measure supported on visited state-action pairs and reconstructs an action-value estimate via kernel integration. The method jointly estimates the stationary distribution of the behavior chain and the Q-measure through coupled stochastic approximation, leading to an efficient weight-based implementation with $O(n)$ memory and $O(n)$ computation cost per iteration. Under uniform ergodicity of the behavior chain, we prove almost sure sup-norm convergence of the induced Q-function to the fixed point of a kernel-smoothed Bellman operator. We also bound the approximation error between this limit and the optimal $Q^*$ as a function of the kernel bandwidth. To assess the performance of our proposed algorithm, we conduct RL experiments in a two-item inventory control setting.
format Preprint
id arxiv_https___arxiv_org_abs_2603_03523
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Q-Measure-Learning for Continuous State RL: Efficient Implementation and Convergence
Wang, Shengbo
Machine Learning
Optimization and Control
We study reinforcement learning in infinite-horizon discounted Markov decision processes with continuous state spaces, where data are generated online from a single trajectory under a Markovian behavior policy. To avoid maintaining an infinite-dimensional, function-valued estimate, we propose the novel Q-Measure-Learning, which learns a signed empirical measure supported on visited state-action pairs and reconstructs an action-value estimate via kernel integration. The method jointly estimates the stationary distribution of the behavior chain and the Q-measure through coupled stochastic approximation, leading to an efficient weight-based implementation with $O(n)$ memory and $O(n)$ computation cost per iteration. Under uniform ergodicity of the behavior chain, we prove almost sure sup-norm convergence of the induced Q-function to the fixed point of a kernel-smoothed Bellman operator. We also bound the approximation error between this limit and the optimal $Q^*$ as a function of the kernel bandwidth. To assess the performance of our proposed algorithm, we conduct RL experiments in a two-item inventory control setting.
title Q-Measure-Learning for Continuous State RL: Efficient Implementation and Convergence
topic Machine Learning
Optimization and Control
url https://arxiv.org/abs/2603.03523