Skip to content

Detail of publication

Citation

Hering, P. : Parameter estimation of MLP network by particle filters . Process control 2006, p. 1-7, Univerzita Pardubice, Pardubice, 2006.

Abstract

Parameter estimation of a multi-layer perceptron network using global nonlinear filtering methods is treated. The truncated second-order filter is applied within the framework of the particle filter approach to design the truncated second-order particle filter as less computationally demanding alternative to unscented Kalman particle filter if it is used for the parameter estimation of multi-layer perceptron network. Further, the particle filter based estimators such as the extended Kalman particle filter and the truncated second-order particle filter are compared in a numerical example.

Detail of publication

Title: Parameter estimation of MLP network by particle filters
Author: Hering, P.
Language: English
Date of publication: 13 Jun 2006
Year: 2006
Type of publication: Papers in proceedings of reviewed conferences
Title of journal or book: Process control 2006
Page: 1 - 7
ISBN: 80-7194-860-8
Publisher: Univerzita Pardubice
Address: Pardubice
Date: 13 Jun 2006 - 16 Jun 2006
/ /

Keywords

Parameter estimation, Probability density function, Particle filters, Multilayer perceptron network

BibTeX

@INPROCEEDINGS{HeringP_2006_Parameterestimation,
 author = {Hering, P.},
 title = {Parameter estimation of MLP network by particle filters},
 year = {2006},
 publisher = {Univerzita Pardubice},
 journal = {Process control 2006},
 address = {Pardubice},
 pages = {1-7},
 ISBN = {80-7194-860-8},
 url = {http://www.kky.zcu.cz/en/publications/HeringP_2006_Parameterestimation},
}