Pattern classification with evolved RBF Neural Networks

This paper shows the latest results of the EvRBF algorithm applied to classification problems. EvRBF is an Evolutionary Algorithm intended to automatically design Radial Basis Functions Neural Networks (RBFNN). EvRBF has been programmed using Evolutionary Objects (EO) library, so that it works with data structures that represent the net directly, instead of binary codifications. This allows the use of specific operators that interchange information between individuals, and produce new ones, relying on the specific features of RBFNN. Experiments using machine learning benchmarks show good behavior of EvRBF with respect to the quality and size of the solutions and the execution time, despite being an Evolutionary Algorithm that does not perform local tuning. These results are compared to other evolutionary neural net algorithms in the literature, showing some improvement over them.

[1]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[2]  D. Broomhead,et al.  Radial Basis Functions, Multi-Variable Functional Interpolation and Adaptive Networks , 1988 .

[3]  John Moody,et al.  Fast Learning in Networks of Locally-Tuned Processing Units , 1989, Neural Computation.

[4]  John C. Platt.  A Resource-Allocating Network for Function Interpolation , 1991, Neural Computation.

[5]  Bernhard E. Boser,et al.  A training algorithm for optimal margin classifiers , 1992, COLT '92.

[6]  Bernd Fritzke.  Supervised Learning with Growing Cell Structures , 1993, NIPS.

[7]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[8]  M. Hoher,et al.  Similarities of LVQ and RBF learning-a survey of learning rules and the application to the classification of signals from high-resolution electrocardiography , 1994, Proceedings of IEEE International Conference on Systems, Man and Cybernetics.

[9]  Lutz Prechelt,et al.  PROBEN 1 - a set of benchmarks and benchmarking rules for neural network training algorithms , 1994 .

[10]  Asim Roy,et al.  An algorithm to generate radial basis function (RBF)-like nets for classification problems , 1995, Neural Networks.

[11]  Brian Carse,et al.  Fast Evolutionary Learning of Minimal Radial Basis Function Neural Networks Using a Genetic Algorithm , 1996, Evolutionary Computing, AISB Workshop.

[12]  Bruce A. Whitehead,et al.  Cooperative-competitive genetic evolution of radial basis function centers and widths for time series prediction , 1996, IEEE Trans. Neural Networks.

[13]  Sheng Chen,et al.  Regularized orthogonal least squares algorithm for constructing radial basis function networks , 1996 .

[14]  Christophe Giraud-Carrier,et al.  GA-RBF: A Self-Optimising RBF Network , 1997, ICANNGA.

[15]  Horst Bischof,et al.  An efficient MDL-based construction of RBF networks , 1998, Neural Networks.

[16]  M. Kubat,et al.  Decision trees can initialize radial-basis function networks , 1998, IEEE Trans. Neural Networks.

[17]  Nathan Intrator,et al.  Global Optimization of RBF Networks , 2000 .

[18]  Juan Julián Merelo Guervós,et al.  G-Prop: Global optimization of multilayer perceptrons using GAs , 2000, Neurocomputing.

[19]  Friedhelm Schwenker,et al.  Three learning phases for radial-basis-function networks , 2001, Neural Networks.

[20]  M. Keijzer,et al.  Evolving Objects: A General Purpose Evolutionary Computation Library , 2001, Artificial Evolution.

[21]  Héctor Pomares,et al.  Expert Mutation Operators for the Evolution of Radial Basis Function Neural Networks , 2001, IWANN.

[22]  William H. Press,et al.  Numerical recipes in C , 2002 .

[23]  Victor Manuel Pérez Santos,et al.  Optimización de redes neuronales de funciones base radiales mediante algoritmos evolutivos , 2003 .

[24]  Pedro Ángel Castillo Valdivieso,et al.  Optimización de perceptrones multicapa mediante algoritmos evolutivos , 2006 .

[25]  Moshe Sipper,et al.  Evolving artificial neural networks with FINCH , 2013, GECCO '13 Companion.

[26]  Tsuyoshi Murata,et al.  {m , 1934, ACML.