Dimensionality reduction in conic section function neural network

This paper details how dimensionality can be reduced in conic section function neural networks (CSFNN). This is particularly important for hardware implementation of networks. One of the main problems to be solved when considering the hardware design is the high connectivity requirement. If the effect that each of the network inputs has on the network output after training a neural network is known, then some inputs can be removed from the network. Consequently, the dimensionality of the network, and hence, the connectivity and the training time can be reduced. Sensitivity analysis, which extracts the cause and effect relationship between the inputs and outputs of the network, has been proposed as a method to achieve this and is investigated for Iris plant, thyroid disease and ionosphere databases. Simulations demonstrate the validity of the method used.

[1]  P N Chatur,et al.  Thyroid Disease Recognition Using Artificial Neural Networks , 2000 .

[2]  S. Hashem,et al.  Sensitivity analysis for feedforward artificial neural networks with differentiable activation functions , 1992, [Proceedings 1992] IJCNN International Joint Conference on Neural Networks.

[3]  David S. Broomhead,et al.  Multivariable Functional Interpolation and Adaptive Networks , 1988, Complex Syst..

[4]  Kevin Baker,et al.  Classification of radar returns from the ionosphere using neural networks , 1989 .

[5]  F. Girosi,et al.  Networks for approximation and learning , 1990, Proc. IEEE.

[6]  Mohamed S. Kamel,et al.  Minimum Complexity Neural Networks for Classification , 1994 .

[7]  Tulay Yildirim,et al.  A unified framework for connectionist models , 1997, NCPW.

[8]  Mohamed I. Elmasry,et al.  VLSI Artificial Neural Networks Engineering , 1994 .

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

[10]  Ulrich Rückert,et al.  VLSI Design of Neural Networks , 1990 .

[11]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[12]  Norman E. Sondak,et al.  Neural networks and artificial intelligence , 1989, SIGCSE '89.

[13]  José G. Delgado-Frias,et al.  VLSI for Neural Networks and Artificial Intelligence , 1994, Springer US.

[14]  R. Fisher.  THE USE OF MULTIPLE MEASUREMENTS IN TAXONOMIC PROBLEMS , 1936 .

[15]  G. Dorffner.  UNIFIED FRAMEWORK FOR MLPs AND RBFNs: INTRODUCING CONIC SECTION FUNCTION NETWORKS , 1994 .