Sample Complexity of Variance-reduced Distributionally Robust Q-learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Shengbo, Si, Nian, Blanchet, Jose, Zhou, Zhengyuan
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912013822722048
author Wang, Shengbo
Si, Nian
Blanchet, Jose
Zhou, Zhengyuan
author_facet Wang, Shengbo
Si, Nian
Blanchet, Jose
Zhou, Zhengyuan
contents Dynamic decision-making under distributional shifts is of fundamental interest in theory and applications of reinforcement learning: The distribution of the environment in which the data is collected can differ from that of the environment in which the model is deployed. This paper presents two novel model-free algorithms, namely the distributionally robust Q-learning and its variance-reduced counterpart, that can effectively learn a robust policy despite distributional shifts. These algorithms are designed to efficiently approximate the $q$-function of an infinite-horizon $γ$-discounted robust Markov decision process with Kullback-Leibler ambiguity set to an entry-wise $ε$-degree of precision. Further, the variance-reduced distributionally robust Q-learning combines the synchronous Q-learning with variance-reduction techniques to enhance its performance. Consequently, we establish that it attains a minimax sample complexity upper bound of $\tilde O(|\mathbf{S}||\mathbf{A}|(1-γ)^{-4}ε^{-2})$, where $\mathbf{S}$ and $\mathbf{A}$ denote the state and action spaces. This is the first complexity result that is independent of the ambiguity size $δ$, thereby providing new complexity theoretic insights. Additionally, a series of numerical experiments confirm the theoretical findings and the efficiency of the algorithms in handling distributional shifts.
format Preprint
id arxiv_https___arxiv_org_abs_2305_18420
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Sample Complexity of Variance-reduced Distributionally Robust Q-learning
Wang, Shengbo
Si, Nian
Blanchet, Jose
Zhou, Zhengyuan
Machine Learning
Optimization and Control
Dynamic decision-making under distributional shifts is of fundamental interest in theory and applications of reinforcement learning: The distribution of the environment in which the data is collected can differ from that of the environment in which the model is deployed. This paper presents two novel model-free algorithms, namely the distributionally robust Q-learning and its variance-reduced counterpart, that can effectively learn a robust policy despite distributional shifts. These algorithms are designed to efficiently approximate the $q$-function of an infinite-horizon $γ$-discounted robust Markov decision process with Kullback-Leibler ambiguity set to an entry-wise $ε$-degree of precision. Further, the variance-reduced distributionally robust Q-learning combines the synchronous Q-learning with variance-reduction techniques to enhance its performance. Consequently, we establish that it attains a minimax sample complexity upper bound of $\tilde O(|\mathbf{S}||\mathbf{A}|(1-γ)^{-4}ε^{-2})$, where $\mathbf{S}$ and $\mathbf{A}$ denote the state and action spaces. This is the first complexity result that is independent of the ambiguity size $δ$, thereby providing new complexity theoretic insights. Additionally, a series of numerical experiments confirm the theoretical findings and the efficiency of the algorithms in handling distributional shifts.
title Sample Complexity of Variance-reduced Distributionally Robust Q-learning
topic Machine Learning
Optimization and Control
url https://arxiv.org/abs/2305.18420