Combined projection and kernel basis functions for classification in evolutionary neural

This paper proposes a hybrid neural network model using a possible combination of different transfer projection functions (sigmoidal unit, SU, product unit, PU) and kernel functions (radial basis function, RBF) in the hidden layer of a feed-forward neural network. An evolutionary algorithm is adapte...

Descripción completa

Detalles Bibliográficos
Autores: Gutiérrez Peña, Pedro Antonio, Hervás Martínez, César, Carbonero Ruz, Mariano, Fernández Caballero, Juan Carlos
Tipo de recurso: artículo
Fecha de publicación:2008
País:España
Institución:Universidad Loyola Andalucía
Repositorio:Brújula
OAI Identifier:oai:repositorio.uloyola.es:20.500.12412/1025
Acceso en línea:http://hdl.handle.net/20.500.12412/1025
Access Level:acceso abierto
Palabra clave:Projection basis functions
Kernel basis functions
Evolutionary neural networks
Descripción
Sumario:This paper proposes a hybrid neural network model using a possible combination of different transfer projection functions (sigmoidal unit, SU, product unit, PU) and kernel functions (radial basis function, RBF) in the hidden layer of a feed-forward neural network. An evolutionary algorithm is adapted to this model and applied for learning the architecture, weights and node typology. Three different combined basis function models are proposed with all the different pairs that can be obtained with SU, PU and RBF nodes: product–sigmoidal unit (PSU) neural networks, product–radial basis function (PRBF) neural networks, and sigmoidal–radial basis function (SRBF) neural networks; and these are compared to the corresponding pure models: product unit neural network (PUNN), multilayer perceptron (MLP) and the RBF neural network.