Skip to content

Publications

Detail of publication

Citation

Hering Pavel and Šimandl Miroslav : Sequential optimal experiment design for neural networks using multiple linearization . Neurocomputing, vol. 73, p. 3284-3290, Elsevier, Amsterdam, 2010.

Abstract

Design of an optimal input signal in system identification using multi-layer perceptron network is treated. It is shown that utilizing the conditional probability density function of parameters for design of the input signal provides better results than currently used procedures based on prameter point estimates only. The conditional probability density function of parameters is approximated by a sum of normal distributions.

Detail of publication

Title: Sequential optimal experiment design for neural networks using multiple linearization
Author: Hering Pavel ; Šimandl Miroslav
Language: English
Year: 2010
Type of publication: Papers in journals
Title of journal or book: Neurocomputing
Číslo vydání: 73
Page: 3284 - 3290
ISSN: 0925-2312
Publisher: Elsevier
Address: Amsterdam
2011-03-15 16:21:55 / 2011-03-15 16:21:55 / 1

Keywords

System identification, optimal experiment design, nonlinear parameter estimation, multi-layer perceptron network

BibTeX

@ARTICLE{HeringPavel_2010_Sequentialoptimal,
 author = {Hering Pavel and \v{S}imandl Miroslav},
 title = {Sequential optimal experiment design for neural networks using multiple linearization},
 year = {2010},
 publisher = {Elsevier},
 journal = {Neurocomputing},
 address = {Amsterdam},
 volume = {73},
 pages = {3284-3290},
 ISSN = {0925-2312},
 url = {http://www.kky.zcu.cz/en/publications/HeringPavel_2010_Sequentialoptimal},
}