Title :
Nonparametric regression estimation by normalized radial basis function networks
Author :
A. Krzyzak;D. Schafer
Author_Institution :
Dept. of Comput. Sci. & Software Eng., Concordia Univ., Montreal, Que., Canada
Abstract :
This paper establishes weak and strong universal consistency of regression estimates based on normalized radial basis function networks when the network parameters are chosen by empirical risk minimization.
Keywords :
"Radial basis function networks","Kernel","Risk management","Multilayer perceptrons","Interpolation","Smoothing methods","Regression analysis","Eigenvalues and eigenfunctions","Neural networks"
Journal_Title :
IEEE Transactions on Information Theory
DOI :
10.1109/TIT.2004.842632
