Abstract
In this paper, neural networks based on orthonormal wavelets are constructed to decompose signals into full scale space. Two algorithms, the global multiscale learning (GML) and the pyramid multiscale learning (PML) were proposed and comparatively studied for their convergence speed, accuracy, and multiresolution representation property in functional approximation. It is shown that the GML algorithm produces larger approximation errors, fails to approximate a chaotic signal, and does not posses multiresolution representation property. On the other hand, the PML algorithm produces smaller approximation errors, can approximate a chaotic signal, and possesses multiresolution representation property.
Chapter PDF
Similar content being viewed by others
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Q. Zhang, “Wavelet networks,” IEEE Trans. Neural Network, vol. 3, no. 6, pp. 889–898, Nov. 1992
T. I. Boubez and R. L. Peskin, “Wavelet neural networks and receptive field partitioning,” IEEE Int. Conf. Neural Networks, vol. 3, pp. 1544–1549, Mar. 1993
Y. C. Pati and P. S. Krishnaprasad, “Analysis and synthesis of feedforward neural networks using discrete affine wavelet transformations,” IEEE Trans. Neural Network, vol. 4, no. 1, pp. 73–85, Jan. 1993
R. Hecht-Nielsen, “Theory of the backpropagation neural network,“ Proc. IJCNN, Washington D. C., I–593, Jun. 18–22, 1989
G. Cybenko, “Approximation by superposition of a sigmoidal function,” Mathematics of Control, Signal and Systems, vol. 2, pp. 303–314, 1989
S. M. Carrol and B. W. Dickinson, “Construction of neural nets using the radon transform,” Proc. IJCNN, 1989
A. R. Webb, “Functional approximation by feedforward networks: a least squares approach to generalization,” IEEE Trans. Neural Networks, vol. 5, no. 3, pp. 363–371, May 1994
I. Daubechies, “Orthonormal bases of compactly supported wavelets,” Comm. on Pure and Applied Math., vol. 91, pp. 909–996, 1988
S. G. Mallat, “A theory for multiresolution signal decomposition: the wavelet representation,” IEEE Trans. PAMI, vol. 11, no. 7, pp. 674–693, Jul. 1989
-, “Multiresolution approximation and wavelets: orthonormal bases of L 2 (R),” Trans. Amer. Math. Soc., 315, no. 1, pp.69–88, 1989
S. G. Mallat and S. Zhong, “Characterization of signals from multiscale edges,” IEEE Trans. PAMI, vol. 14, no. 7, pp. 710–732, Jul. 1992
H. D. I. Abarbanel, “Chaotic signals and physical systems, ” Proc. IEEE ICASSP, 1992
J. C. Luo and C. H. Chen, “A study of signal decomposition using pyramidal wavelet networks,” Master Thesis, National Cheng Kung Univ., Taiwan, R. O. C., Jun. 1995
D. Hush, J. M. Salas, and B. Horne, “Error surfaces for multilayer perceptrons,” IEEE Trans. SMC, vol. 22, no. 5, 1992
B. Delyon, A. Juditsky, and A. Benveniste, “Accuracy analysis for wavelet approximations,” IEEE Trans. Neural Networks, vol. 6, no. 2, pp. 332–348, Mar. 1995
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Wang, J.W., Chen, C.H., Luo, J.C. (1996). Signal decomposition with multiscale learning algorithms. In: Perner, P., Wang, P., Rosenfeld, A. (eds) Advances in Structural and Syntactical Pattern Recognition. SSPR 1996. Lecture Notes in Computer Science, vol 1121. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-61577-6_18
Download citation
DOI: https://doi.org/10.1007/3-540-61577-6_18
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-61577-4
Online ISBN: 978-3-540-70631-1
eBook Packages: Springer Book Archive
