Performance evaluation of two radial basis function neural network models

BHOS Repository

Performance evaluation of two radial basis function neural network models

Show full item record

Title: Performance evaluation of two radial basis function neural network models
Author: Chistova, Nikolinka G; Vachkov, Gancho L
Abstract: In this paper performance evaluation of two modification of the classical Radial Basis Function Neural Network (RBFNN) model, called Reduced and Simplified RBFNN models is carried out. Different RBFNN models with different number of Radial Basis Functions (RBFs) are created and analyzed. Particle Swarm Optimization (PSO) algorithm with constraints for the parameter tuning of the models is applied. Simultaneous optimization of all three groups of parameters, namely the centers, widths and the weights of the RBFNN is performed. It is shown that the Simplified RBFNN models, which have smaller number of parameters, can achieve even better modeling accuracy than the Reduced RBFNN models.
URI: http://hdl.handle.net/123456789/1387
Date: 2014


Files in this item

Files Size Format View
Christova_Vachkov_SYSTEMS2-07.pdf 436.5Kb PDF View/Open

This item appears in the following Collection(s)

Show full item record

Search BHOS Repository


Advanced Search

Browse

My Account