Controlling the Asymptotic Level Density for Quantization Processes with Self-Organizing Maps
We analyze different methods of controlling magnification factor in the relation between the density of the input vectors and the asymptotic density of the quantizers. We also propose a new method of modeling the one-dimensional self-organizing process and examine some properties of the method. Empirical studies show strong influence of the learning rate strategies on the observed magnification factor of the vector quantization process.
KeywordsLearning Rate Magnification Factor Quantization Process Reference Vector Weight Adjustment
Unable to display preview. Download preview PDF.
- 1.H. U. Bauer, R. Der, M. Herman, Controlling the magnification factor of self-organizing feature maps Neural Computation, vol. 8, 757–771, 1996.Google Scholar
- 3.A. Gersho, R.M. Gray, Vector quantization and Signal Compression, Kluwer Academic Publisher, Boston, 1993.Google Scholar
- 4.T. Heskes, B. Kappen, Self-organization and parametric regression, Poc. ICANN95, Paris, France, 81–86, 1995.Google Scholar
- 8.T. Kohonen, Self-organizing maps, Berlin, Heidelberg Springer Verlag (1995).Google Scholar
- 11.T.M. Martinetz, S.G. Berkovich, K.J. Schulten, Neural-gas network for vector quantization and its application to time-series prediction IEEE Trans. on Neural Networks, vol.4., 558–559 (1993).Google Scholar
- 16.E. Skubalska-Rafajlowicz, Space-Filling Curves and Kohonen’s 1-D SOM as a Method of a Vector Quantization with Known Asymptotic Level Density, Proc. of the 3rd Conf. Neural Networks and Their Applications, Kule 1997, pp. 161–166.Google Scholar