Improving the effectiveness of RBF classifier based on a hybrid cost function

This paper presents a new minimum classification error (MCE)–mean square error (MSE) hybrid cost function to enhance the classification ability and speed up the learning process of radial basis function (RBF)-based classifier. Contributed by the MCE function, the proposed cost function enables the RBF-based classifier to achieve an excellent classification performance compared with the conventional MSE function. In addition, certain learning difficulties experienced by the MCE algorithm can be solved in an efficient and simple way. The presented results show that the proposed method exhibits a substantially higher convergence rate compared with the MCE function.

[1]  Biing-Hwang Juang,et al.  Pattern recognition using a family of design algorithms based upon the generalized probabilistic descent method , 1998, Proc. IEEE.

[2]  Shigeru Katagiri,et al.  Discriminative metric design for robust pattern recognition , 1997, IEEE Trans. Signal Process..

[3]  Hilary Buxton,et al.  Learning identity with radial basis function networks , 1998, Neurocomputing.

[4]  S. Hyakin,et al.  Neural Networks: A Comprehensive Foundation , 1994 .

[5]  Tommy W. S. Chow,et al.  Induction machine fault detection using SOM-based RBF neural networks , 2004, IEEE Transactions on Industrial Electronics.

[6]  Geoffrey E. Hinton,et al.  Proceedings of the 1988 Connectionist Models Summer School , 1989 .

[7]  Guoqiang Peter Zhang,et al.  Neural networks for classification: a survey , 2000, IEEE Trans. Syst. Man Cybern. Part C.

[8]  Tommy W. S. Chow,et al.  Induction machine fault diagnostic analysis with wavelet technique , 2004, IEEE Transactions on Industrial Electronics.

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

[10]  Bernhard Schölkopf,et al.  Comparing support vector machines with Gaussian kernels to radial basis function classifiers , 1997, IEEE Trans. Signal Process..

[11]  Michele Ceccarelli,et al.  Sequence recognition with radial basis function networks: experiments with spoken digits , 1996, Neurocomputing.

[12]  Tommy W. S. Chow,et al.  Three phase induction machines asymmetrical faults identification using bispectrum , 1995 .

[13]  Yves Chauvin,et al.  Backpropagation: theory, architectures, and applications , 1995 .

[14]  T Poggio,et al.  Regularization Algorithms for Learning That Are Equivalent to Multilayer Networks , 1990, Science.

[15]  Biing-Hwang Juang,et al.  Discriminative learning for minimum error classification [pattern recognition] , 1992, IEEE Trans. Signal Process..

[16]  Régis Lengellé,et al.  Training MLPs layer by layer using an objective function for internal representations , 1996, Neural Networks.

[17]  T. Poggio,et al.  Networks and the best approximation property , 1990, Biological Cybernetics.

[18]  K. Lang,et al.  Learning to tell two spirals apart , 1988 .

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

[20]  M. V. Velzen,et al.  Self-organizing maps , 2007 .

[21]  Susan B. Garavaglia.  The two spirals benchmark: lessons from the hidden layers , 1999, IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339).

[22]  C. C. Homes,et al.  Bayesian Radial Basis Functions of Variable Dimension , 1998, Neural Computation.

[23]  Christian Lebiere,et al.  The Cascade-Correlation Learning Architecture , 1989, NIPS.

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

[25]  Marco Sciandrone,et al.  Efficient training of RBF neural networks for pattern recognition , 2001, IEEE Trans. Neural Networks.

[26]  Tommy W. S. Chow,et al.  HOS-based nonparametric and parametric methodologies for machine fault detection , 2000, IEEE Trans. Ind. Electron..