Event-driven stochastic approximation

We consider a Robbins-Monro type iteration wherein noisy measurements are event-driven and therefore arrive asynchronously. We propose a modification of step-sizes that ensures desired asymptotic behaviour regardless of this aspect. This generalizes earlier results on asynchronous stochastic approxi...

Full description

Saved in:

Bibliographic Details
Published in:	Indian journal of pure and applied mathematics Vol. 47; no. 2; pp. 291 - 299
Main Authors:	Borkar, Vivek S., Sahasrabudhe, Neeraja, Ashok Vardhan, M.
Format:	Journal Article
Language:	English
Published:	New Delhi Indian National Science Academy 01-06-2016
Subjects:	Applications of Mathematics Mathematics Mathematics and Statistics Numerical Analysis asynchronous computation Stochastic approximation o.d.e. limit sampling bias distributed algorithms
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	We consider a Robbins-Monro type iteration wherein noisy measurements are event-driven and therefore arrive asynchronously. We propose a modification of step-sizes that ensures desired asymptotic behaviour regardless of this aspect. This generalizes earlier results on asynchronous stochastic approximation wherein the asynchronous behaviour is across different components, but not along the same component of the vector iteration, as is the case considered here.
ISSN:	0019-5588 0975-7465
DOI:	10.1007/s13226-016-0188-1