FERNÁNDEZ NAVARRO, FRANCISCO DE ASÍS, Hervás-Martínez C. , Gutiérrez P.A. , Cruz-Ramírez M. , CARBONERO RUZ, MARIANO

No

Lect. Notes Comput. Sci.

Conference Paper

Científica

✗

✗

0.322

https://www.scopus.com/inward/record.uri?eid=2-s2.0-77954602192&doi=10.1007%2f978-3-642-13803-4_35&partnerID=40&md5=847af9394dd7e1fbf3c1369a8d799822

01/01/2010

000286905700035

2-s2.0-77954602192

10.1007/978-3-642-13803-4_35

This paper proposes a Radial Basis Function Neural Network (RBFNN) which reproduces different Radial Basis Functions (RBFs) by means a real parameter q, named q-Gaussian RBFNN. The architecture, weights and node topology are learnt through a Hybrid Algorithm (HA) with the iRprop?+ algorithm as the local improvement procedure. In order to test its overall performance, an experimental study with eleven datasets, taken from the UCI repository is presented. The RBFNN with the q-Gaussian is compared to RBFNN with Gaussian, Cauchy and Inverse Multiquadratic RBFs. © 2010 Springer-Verlag.

Data sets; Experimental studies; Gaussian radial basis functions; Gaussians; Hybrid algorithms; Multiquadratics; Radial basis function neural networks; Radial basis functions; UCI repository; Gaussian distribution; Image segmentation; Radial basis function networks; Neural networks

• MARIANO CARBONERO RUZ