Saved in:
Bibliographic Details
Main Authors: Wang, Shengbo, Zhang, Zexi
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.08417
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914546045681664
author Wang, Shengbo
Zhang, Zexi
author_facet Wang, Shengbo
Zhang, Zexi
contents Designing model-free algorithms for distributionally robust reinforcement learning (DRRL) poses fundamental challenges. The robust Bellman operator is nonlinear in the transition kernel, which makes one-sample Bellman updates biased, while the adversarial optimization underlying robustness makes robust evaluation computationally demanding. To address these difficulties, we consider the natural small-ambiguity regime under Kullback--Leibler ambiguity sets and propose an approximate DRRL framework based on a first-order expansion of the relevant robust functional. This yields an approximate robust Bellman equation that removes the adversarial optimization while remaining first-order accurate in the ambiguity radius. To learn the fixed point of this approximate equation, we propose Mean-Variance Stochastic Approximation (MVSA), a model-free algorithm that uses only one-sample updates. This is achieved via a lifted stochastic approximation dynamics and a two-time-scale design. We then prove convergence and a central limit theorem for MVSA: its main iterate satisfies a central limit theorem at the canonical $n^{-1/2}$ scale, with explicitly characterized asymptotic covariances. Finally, we validate our theoretical findings with a numerical experiment.
format Preprint
id arxiv_https___arxiv_org_abs_2605_08417
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Central Limit Theorem for Two-Time-Scale Approximate Distributionally Robust RL
Wang, Shengbo
Zhang, Zexi
Machine Learning
Optimization and Control
Designing model-free algorithms for distributionally robust reinforcement learning (DRRL) poses fundamental challenges. The robust Bellman operator is nonlinear in the transition kernel, which makes one-sample Bellman updates biased, while the adversarial optimization underlying robustness makes robust evaluation computationally demanding. To address these difficulties, we consider the natural small-ambiguity regime under Kullback--Leibler ambiguity sets and propose an approximate DRRL framework based on a first-order expansion of the relevant robust functional. This yields an approximate robust Bellman equation that removes the adversarial optimization while remaining first-order accurate in the ambiguity radius. To learn the fixed point of this approximate equation, we propose Mean-Variance Stochastic Approximation (MVSA), a model-free algorithm that uses only one-sample updates. This is achieved via a lifted stochastic approximation dynamics and a two-time-scale design. We then prove convergence and a central limit theorem for MVSA: its main iterate satisfies a central limit theorem at the canonical $n^{-1/2}$ scale, with explicitly characterized asymptotic covariances. Finally, we validate our theoretical findings with a numerical experiment.
title Central Limit Theorem for Two-Time-Scale Approximate Distributionally Robust RL
topic Machine Learning
Optimization and Control
url https://arxiv.org/abs/2605.08417