Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Spectrumbased Design of Sinusoidal RBF Neural Networks Pter Andrs
 

Summary: Spectrum­based Design of Sinusoidal RBF Neural Networks
Péter András
Department of Psychology
University of Newcastle
Newcastle upon Tyne, NE1 7RU, UK
peter.andras@ncl.ac.uk
Abstract
The paper introduces and describes the spectrum­based
design of RBF neural networks. The RBF neural networks
used in the paper work with damped sinusoidal nonlinear
activation functions. The concept of the associated spectrum
of the data is introduced and it is shown how to apply this
spectrum to find the number and internal parameters of
hidden neurons for a neural network solution of the related
data processing problem. A time series prediction
application is presented. The relation of the proposed
method to the support vector machine method and the
application of the method to select appropriate basis
functions for a problem with given data are discussed.
I. INTRODUCTION

  

Source: Andras, Peter - School of Computing Science, University of Newcastle upon Tyne

 

Collections: Computer Technologies and Information Sciences