Skip to content

Detail of publication

Citation

Hering, P. and Král, L. and Šimandl, M. : Local nonlinear filters with the second-order accuracy for neural network parameter estimation . Proceedings of 7th International Carpathian control conference, p. 165-168, VŠB -Technická univerzita, Ostrava, 2006.

Abstract

This paper concentrates on parameter estimation of multi-layer perceptron network by nonlinear filtering methods. Whereas the EKF is using pproximation by the first order Taylor series expansion, the UKF estimates the parameters numerically by using a set of deterministically chosen points having higher degree of accuracy than the EKF without need to compute the derivatives of the function describing the network. Comparable accuracy as by UKF could be obtained by the truncated second-order filter (TSF). TSF has not been used yet because it is considered as more computationally demanding due to need of evaluation the matrix of second order derivatives. This paper points out, that the SKF is faster than the UKF in parameter estimation of MLP network thanks to a special form of the matrix of the second order derivatives. The evaluation of estimation quality and computational demands are demonstrated by a numerical example.

Detail of publication

Title: Local nonlinear filters with the second-order accuracy for neural network parameter estimation
Author: Hering, P. ; Král, L. ; Šimandl, M.
Language: English
Date of publication: 29 May 2006
Year: 2006
Type of publication: Papers in proceedings of reviewed conferences
Title of journal or book: Proceedings of 7th International Carpathian control conference
Page: 165 - 168
ISBN: 80-248-1066-2
Publisher: VŠB -Technická univerzita
Address: Ostrava
Date: 29 May 2006 - 31 May 2006
/ /

Keywords

Parameter estimation, local filters, accuracy, multi-layer perceptron network

BibTeX

@INPROCEEDINGS{HeringP_2006_Localnonlinear,
 author = {Hering, P. and Kr\'{a}l, L. and \v{S}imandl, M.},
 title = {Local nonlinear filters with the second-order accuracy for neural network parameter estimation},
 year = {2006},
 publisher = {V\v{S}B -Technick\'{a} univerzita},
 journal = {Proceedings of 7th International Carpathian control conference},
 address = {Ostrava},
 pages = {165-168},
 ISBN = {80-248-1066-2},
 url = {http://www.kky.zcu.cz/en/publications/HeringP_2006_Localnonlinear},
}