Non-Asymptotic Convergence of Stochastic Iterative Algorithms: A Lyapunov Framework

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Zaiwei, Maguluri, Siva Theja
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917548044320768
author Chen, Zaiwei
Maguluri, Siva Theja
author_facet Chen, Zaiwei
Maguluri, Siva Theja
contents We survey Lyapunov-based techniques for the finite-time analysis of stochastic iterative algorithms, also known as stochastic approximation (SA) algorithms, for solving fixed-point equations $\bar{F}(x)=x$, where the operator $\bar{F}(\cdot)$ can only be accessed through a noisy oracle. We first focus on the standard setting in which $\bar{F}(\cdot)$ is contractive with respect to some norm and the noise is i.i.d., and explain how generalized Moreau envelopes serve as universal Lyapunov functions, regardless of the underlying norm. We then show how this framework yields mean-square convergence guarantees and applies to stochastic gradient descent, linear SA, and value-based reinforcement learning algorithms such as Q-learning and temporal-difference learning. Finally, we discuss extensions to Markovian noise, seminorm-contractive operators, dissipative operators, and high-probability bounds, and conclude with open problems. The goal is to present a unified and self-contained roadmap for the finite-time analysis of SA and its applications, especially in reinforcement learning.
format Preprint
id arxiv_https___arxiv_org_abs_2605_31309
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Non-Asymptotic Convergence of Stochastic Iterative Algorithms: A Lyapunov Framework
Chen, Zaiwei
Maguluri, Siva Theja
Machine Learning
Probability
We survey Lyapunov-based techniques for the finite-time analysis of stochastic iterative algorithms, also known as stochastic approximation (SA) algorithms, for solving fixed-point equations $\bar{F}(x)=x$, where the operator $\bar{F}(\cdot)$ can only be accessed through a noisy oracle. We first focus on the standard setting in which $\bar{F}(\cdot)$ is contractive with respect to some norm and the noise is i.i.d., and explain how generalized Moreau envelopes serve as universal Lyapunov functions, regardless of the underlying norm. We then show how this framework yields mean-square convergence guarantees and applies to stochastic gradient descent, linear SA, and value-based reinforcement learning algorithms such as Q-learning and temporal-difference learning. Finally, we discuss extensions to Markovian noise, seminorm-contractive operators, dissipative operators, and high-probability bounds, and conclude with open problems. The goal is to present a unified and self-contained roadmap for the finite-time analysis of SA and its applications, especially in reinforcement learning.
title Non-Asymptotic Convergence of Stochastic Iterative Algorithms: A Lyapunov Framework
topic Machine Learning
Probability
url https://arxiv.org/abs/2605.31309