Transactions of the Institute of Systems, Control and Information Engineers
Online ISSN : 2185-811X
Print ISSN : 1342-5668
ISSN-L : 1342-5668
Papers
Stopping Rules for Linear Stochastic Approximation
Takayuki WadaTakamitsu ItaniYasumasa Fujisaki
Author information
JOURNALS FREE ACCESS

2012 Volume 25 Issue 2 Pages 39-44

Details
Abstract

Stopping rules are developed for stochastic approximation which is an iterative method for solving an unknown equation based on its consecutive residuals corrupted by additive random noise. It is assumed that the equation is linear and the noise is independent and identically distributed random vectors with zero mean and a bounded covariance. Then, the number of iterations for achieving a given probabilistic accuracy of the resultant solution is derived, which gives a rigorous stopping rule for the stochastic approximation. This number is polynomial of the problem size.

Information related to the author
© 2012 The Institute of Systems, Control and Information Engineers
Previous article
feedback
Top