Standard

Optimal rate of convergence for randomized algorithms of stochastic approximation under arbitrary perturbations. / Granichin, O. N.

в: Avtomatika i Telemekhanika, № 2, 01.01.2003, стр. 88-99.

Результаты исследований: Научные публикации в периодических изданияхстатьяРецензирование

Harvard

APA

Vancouver

Author

BibTeX

@article{a66bea033df4480681ad68d3cfed5ecf,
title = "Optimal rate of convergence for randomized algorithms of stochastic approximation under arbitrary perturbations",
abstract = "Multidimensional stochastic optimization plays important role in analysis and control of many technical systems. For solving difficult multidimensional optimization problems, it is suggested to use randomized stochastic approximation algorithms with input perturbation. These algorithms have simple form and give reliable estimates of unknown parameters at 'almost arbitrary' interference in observations. The optimal techniques of algorithm parameters selection are substantiated.",
author = "Granichin, {O. N.}",
year = "2003",
month = jan,
day = "1",
language = "русский",
pages = "88--99",
journal = "АВТОМАТИКА И ТЕЛЕМЕХАНИКА",
issn = "0005-2310",
publisher = "Издательство {"}Наука{"}",
number = "2",

}

RIS

TY - JOUR

T1 - Optimal rate of convergence for randomized algorithms of stochastic approximation under arbitrary perturbations

AU - Granichin, O. N.

PY - 2003/1/1

Y1 - 2003/1/1

N2 - Multidimensional stochastic optimization plays important role in analysis and control of many technical systems. For solving difficult multidimensional optimization problems, it is suggested to use randomized stochastic approximation algorithms with input perturbation. These algorithms have simple form and give reliable estimates of unknown parameters at 'almost arbitrary' interference in observations. The optimal techniques of algorithm parameters selection are substantiated.

AB - Multidimensional stochastic optimization plays important role in analysis and control of many technical systems. For solving difficult multidimensional optimization problems, it is suggested to use randomized stochastic approximation algorithms with input perturbation. These algorithms have simple form and give reliable estimates of unknown parameters at 'almost arbitrary' interference in observations. The optimal techniques of algorithm parameters selection are substantiated.

UR - http://www.scopus.com/inward/record.url?scp=0037258022&partnerID=8YFLogxK

M3 - статья

AN - SCOPUS:0037258022

SP - 88

EP - 99

JO - АВТОМАТИКА И ТЕЛЕМЕХАНИКА

JF - АВТОМАТИКА И ТЕЛЕМЕХАНИКА

SN - 0005-2310

IS - 2

ER -

ID: 32480737