Comparative Study of Different Types of Wavelet Functions in Neural Network

Based on the wavelet transform theory, the new notion of the wavelet network is proposed as an alternative to feed forward neural networks for approximating arbitrary nonlinear functions. Boubez employed ortho-normal wavelets; Yamakawa, Uchino and Samatsu proposed two types of new neuron models and named Wavelet Synapse (WS) neuron and Wavelet Activation (WA) function neuron. These models are obtained by modifying the Mc Culloch and Pitts neuron model with non-orthogonal wavelet bases. Comparative study of different type of Wavelet functions is carried out in this paper by applying above two neuron models.

[1]  J. Freidman,et al.  Multivariate adaptive regression splines , 1991 .

[2]  Amara Lynn Graps,et al.  An introduction to wavelets , 1995 .

[3]  Jun Zhang,et al.  Wavelet neural networks for function learning , 1995, IEEE Trans. Signal Process..

[4]  Cheng-Jian Lin,et al.  A Wavelet-Based Neuro-Fuzzy System and its Applications , 2007, Intell. Autom. Soft Comput..

[5]  Stephen H. Lane,et al.  Multi-Layer Perceptrons with B-Spline Receptive Field Functions , 1990, NIPS.

[6]  Richard S. Sutton,et al.  Iterative Construction of Sparse Polynomial Approximations , 1991, NIPS.

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

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

[9]  T. Sanger.  Basis-Function Trees for Approximation in High-Dimensional Spaces , 1991 .

[10]  Daniel W. C. Ho,et al.  Fuzzy wavelet networks for function learning , 2001, IEEE Trans. Fuzzy Syst..

[11]  R. L. Peskin,et al.  Wavelet neural networks and receptive field partitioning , 1993, IEEE International Conference on Neural Networks.

[12]  Takeshi Yamakawa,et al.  Wavelet neural networks employing over-complete number of compactly supported non-orthogonal wavelets and their applications , 1994, Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).

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