A Note on Stability in Asynchronous Stochastic Approximation without Communication Delays

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Yu, Huizhen, Wan, Yi, Sutton, Richard S.
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913467592605696
author Yu, Huizhen
Wan, Yi
Sutton, Richard S.
author_facet Yu, Huizhen
Wan, Yi
Sutton, Richard S.
contents In this paper, we study asynchronous stochastic approximation algorithms without communication delays. Our main contribution is a stability proof for these algorithms that extends a method of Borkar and Meyn by accommodating more general noise conditions. We also derive convergence results from this stability result and discuss their application in important average-reward reinforcement learning problems.
format Preprint
id arxiv_https___arxiv_org_abs_2312_15091
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle A Note on Stability in Asynchronous Stochastic Approximation without Communication Delays
Yu, Huizhen
Wan, Yi
Sutton, Richard S.
Machine Learning
Optimization and Control
62L20 (Primary) 93E35, 90C40 (Secondary)
In this paper, we study asynchronous stochastic approximation algorithms without communication delays. Our main contribution is a stability proof for these algorithms that extends a method of Borkar and Meyn by accommodating more general noise conditions. We also derive convergence results from this stability result and discuss their application in important average-reward reinforcement learning problems.
title A Note on Stability in Asynchronous Stochastic Approximation without Communication Delays
topic Machine Learning
Optimization and Control
62L20 (Primary) 93E35, 90C40 (Secondary)
url https://arxiv.org/abs/2312.15091