Helge Ritter, Klaus Obermayer, Klaus Schulten, and Jeanne Rubner.
Self-organizing maps and adaptive filters.
In J. Leo van Hemmen, Eytan Domany, and Klaus Schulten, editors,
Models of Neural Networks, Physics of Neural Networks, pp. 281-306.
Springer-Verlag, New York, 1991.
RITT91
Topographically organized maps and adaptive filters serve important roles for information processing in the brain and are also promising to facilitate tasks in digital information processing. In this contribution, we report results on two important network models. A first network model comprises the "self-organizing feature maps" of Kohonen. We discuss their relation to optimal representation of data, present results of a mathematical analysis of their behavior near a stationary state, demonstrate the formation of "striped projections", if higher-dimensional feature spaces are to be mapped onto a two-dimensional cortical surface, and present recent simulation results for the somatosensory map of the skin surface and the retinal map in the visual cortex. The second network model is a hierarchical network for principal component analysis. Such a network, when trained with correlated random patterns, develops cells the receptive fields of which correspond to Gabor filters and resemble the receptive fields of "simple cells" in the visual cortex.
Download Full Text
The manuscripts available on our site are provided for your personal
use only and may not be retransmitted or redistributed without written
permissions from the paper's publisher and author. You may not upload any
of this site's material to any public server, on-line service, network, or
bulletin board without prior written permission from the publisher and
author. You may not make copies for any commercial purpose. Reproduction
or storage of materials retrieved from this web site is subject to the
U.S. Copyright Act of 1976, Title 17 U.S.C.
