Exit Time Analysis For Kesten's Stochastic Recurrence Equations

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Rhee, Chang-Han, Ryu, Jeeho, Seo, Insuk
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909529297387520
author Rhee, Chang-Han
Ryu, Jeeho
Seo, Insuk
author_facet Rhee, Chang-Han
Ryu, Jeeho
Seo, Insuk
contents Kesten's stochastic recurrent equation is a classical subject of research in probability theory and its applications. Recently, it has garnered attention as a model for stochastic gradient descent with a quadratic objective function and the emergence of heavy-tailed dynamics in machine learning. This context calls for analysis of its asymptotic behavior under both negative and positive Lyapunov exponents. This paper studies the exit times of the Kesten's stochastic recurrence equation in both cases. Depending on the sign of Lyapunov exponent, the exit time scales either polynomially or logarithmically as the radius of the exit boundary increases.
format Preprint
id arxiv_https___arxiv_org_abs_2503_05219
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Exit Time Analysis For Kesten's Stochastic Recurrence Equations
Rhee, Chang-Han
Ryu, Jeeho
Seo, Insuk
Probability
Kesten's stochastic recurrent equation is a classical subject of research in probability theory and its applications. Recently, it has garnered attention as a model for stochastic gradient descent with a quadratic objective function and the emergence of heavy-tailed dynamics in machine learning. This context calls for analysis of its asymptotic behavior under both negative and positive Lyapunov exponents. This paper studies the exit times of the Kesten's stochastic recurrence equation in both cases. Depending on the sign of Lyapunov exponent, the exit time scales either polynomially or logarithmically as the radius of the exit boundary increases.
title Exit Time Analysis For Kesten's Stochastic Recurrence Equations
topic Probability
url https://arxiv.org/abs/2503.05219