A stochastic trained neural network for nonparametric hypothesis testing

The present work is oriented to the description and evaluation as a nonparametric hypothesis test of Genetic Inside Neural Network (GINN) based on an artificial neural network trained with a stochastic process. In this paper, the theoretical framework is detailed, together with the convergence properties, and the way it is implemented. Finally, some studies with simulated and real data show the stability of results and prediction capacity.

[1]  M. S. Sánchez,et al.  Efficiency of multi-layered feed-forward neural networks on classification in relation to linear discriminant analysis, quadratic discriminant analysis and regularized discriminant analysis , 1995 .

[2]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1992, Math. Control. Signals Syst..

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

[4]  M. Bos,et al.  Comparison of the training of neural networks for quantitative x-ray fluorescence spectrometry by a genetic algorithm and backward error propagation , 1991 .

[5]  Halbert White,et al.  Learning in Artificial Neural Networks: A Statistical Perspective , 1989, Neural Computation.

[6]  Kurt Hornik,et al.  Approximation capabilities of multilayer feedforward networks , 1991, Neural Networks.

[7]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[8]  Desire L. Massart,et al.  A non‐parametric class modelling technique , 1989 .

[9]  Lorenzo Bruzzone,et al.  Combination of neural and statistical algorithms for supervised classification of remote-sensing image , 2000, Pattern Recognit. Lett..

[10]  P. Werbos,et al.  Beyond Regression : "New Tools for Prediction and Analysis in the Behavioral Sciences , 1974 .

[11]  Robert M. Burton,et al.  Universal approximation in p-mean by neural networks , 1998, Neural Networks.

[12]  Terrence J. Sejnowski,et al.  Analysis of hidden units in a layered network trained to classify sonar targets , 1988, Neural Networks.

[13]  Riccardo Leardi,et al.  PARVUS: An Extendable Package of Programs for Data Exploration , 1988 .

[14]  Zbigniew Michalewicz,et al.  Genetic Algorithms + Data Structures = Evolution Programs , 1996, Springer Berlin Heidelberg.

[15]  van Km Kees Hee,et al.  A general theory of genetic algorithms , 1989 .

[16]  M. Forina,et al.  Distance and class space in the UNEQ class‐modeling technique , 1995 .

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

[18]  A. Narayanan.  Probability and statistics in engineering and management science , 1972 .

[19]  Luis A. Sarabia,et al.  Typification of vinegars from Jerez and Rioja using classical chemometric techniques and neural network methods , 1999 .

[20]  D. Massart,et al.  UNEQ: a disjoint modelling technique for pattern recognition based on normal distribution , 1986 .

[21]  Bruce R. Kowalski,et al.  Chemometrics, mathematics and statistics in chemistry , 1984 .

[22]  M. S. Sánchez Pastor,et al.  The UNEQ, PLS and MLF neural network methods in the modelling and prediction of the colour of young red wines from the Denomination of Origin ‘Rioja’ , 1995 .

[23]  James A. Freeman,et al.  Simulating neural networks - with Mathematica , 1993 .

[24]  Luis A. Sarabia,et al.  GINN (Genetic Inside Neural Network): towards a non-parametric training , 1997 .

[25]  María Sagrario Sánchez Pastor,et al.  Redes neuronales en clasificación , 1997 .

[26]  Etienne Barnard,et al.  Optimization for training neural nets , 1992, IEEE Trans. Neural Networks.

[27]  L. Jones.  Constructive approximations for neural networks by sigmoidal functions , 1990, Proc. IEEE.

[28]  L.M.C. Buydens,et al.  Performance of multi-layer feedforward and radial base function neural networks in classification and modelling , 1996 .

[29]  Allan Pinkus,et al.  Multilayer Feedforward Networks with a Non-Polynomial Activation Function Can Approximate Any Function , 1991, Neural Networks.