Training of neural classifiers by separating distributions at the hidden layer

Author

Góez, Roger ; Lázaro, Marcelino

Author_Institution

Dept. de Teor. de la Senal y Comun., Univ. Carlos III de Madrid, Leganes, Spain

fYear

2009

fDate

1-4 Sept. 2009

Firstpage

1

Lastpage

6

Abstract

A new cost function for training of binary classifiers based on neural networks is proposed. This cost function aims at separating the distributions for patterns of each class at the output of the hidden layer of the network. It has been implemented in a Generalized Radial Basis Function (GRBF) network and its performance has been evaluated under three different databases, showing advantages with respect to the conventional Mean Squared Error (MSE) cost function. With respect to the Support Vector Machine (SVM) classifier, the proposed method has also advantages both in terms of performance and complexity.

Keywords

learning (artificial intelligence); mean square error methods; pattern classification; radial basis function networks; support vector machines; binary classifiers training; generalized radial basis function network; mean squared error cost function; network hidden layer; neural classifiers training; neural networks; support vector machine classifier; Artificial neural networks; Bayesian methods; Cost function; Curve fitting; Databases; Function approximation; Neural networks; Speech recognition; Support vector machine classification; Support vector machines;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning for Signal Processing, 2009. MLSP 2009. IEEE International Workshop on

Conference_Location

Grenoble

Print_ISBN

978-1-4244-4947-7

Electronic_ISBN

978-1-4244-4948-4

Type

conf

DOI

10.1109/MLSP.2009.5306240

Filename

5306240