Title :
Advantage analysis of sigmoid based RBF networks
Author :
Xing Wu ; Wilamowski, Bogdan M.
Author_Institution :
Dept. of Electr. & Comput. Eng., Auburn Univ., Auburn, AL, USA
Abstract :
By introducing an extra dimension to the inputs, sigmoid function can simulate the behavior of traditional RBF units. This paper introduces a sigmoid based RBF neuron and compares it with traditional RBF neuron. Neural networks composed of these neurons are trained with ErrCor algorithm on two classic experiments. Comparison results are presented to show advantages of the sigmoid based RBF model.
Keywords :
radial basis function networks; ErrCor algorithm; advantage analysis; neural networks; sigmoid based RBF networks; sigmoid based RBF neuron; Algorithm design and analysis; Approximation algorithms; Biological neural networks; Function approximation; Neurons; Radial basis function networks; Training; ErrCor algorithm; RBF; neural network; sigmoid;
Conference_Titel :
Intelligent Engineering Systems (INES), 2013 IEEE 17th International Conference on
Conference_Location :
San Jose
Print_ISBN :
978-1-4799-0828-8
DOI :
10.1109/INES.2013.6632819
