Physiological interpretationm of the self-organizing map algorithm

It is argued that the Self-Organizing Map (SOM) may be implemented in biological neural networks, the cells of which communicate by transsynaptic signals as well as diffuse chemical substances. For the ''Winner Take All'' (WTA) function, a laterally connected neural network seems proper. Hebb's hypothesis about the synaptic modification is replaced in this work by a principle that relates to a chemically interacting small population of neurons. According to this modified law, the synaptic strength vectors also become normalized automatically. The time-variable ''neighborhood function'' needed in the SOM algorithm is most effectively implemented by chemical agents, which are formed or released extracellularly at or in the neighborhood of highly active cells. Such a physiological model then behaves in the same way as the idealized SOM algorithm, which has been found very effective in many information-processing applications.

[1]  C. Malsburg,et al.  How to label nerve cells so that they can interconnect in an ordered fashion. , 1977, Proceedings of the National Academy of Sciences of the United States of America.

[2]  Stephen Grossberg,et al.  Contour Enhancement, Short Term Memory, and Constancies in Reverberating Neural Networks , 1973 .

[3]  Richard Granger,et al.  A cortical model of winner-take-all competition via lateral inhibition , 1992, Neural Networks.

[4]  D. O. Hebb,et al.  The organization of behavior , 1988 .

[5]  Richard P. Lippmann,et al.  An introduction to computing with neural nets , 1987 .

[6]  N. Swindale.  A model for the formation of ocular dominance stripes , 1980, Proceedings of the Royal Society of London. Series B. Biological Sciences.

[7]  T. Kohonen.  Self-organized formation of topographically correct feature maps , 1982 .

[8]  R. Pérez,et al.  Development of Specificity in the Cat Visual Cortex , 1975, Journal of mathematical biology.

[9]  Shun-Ichi Amari,et al.  Topographic organization of nerve fields , 1979, Neuroscience Letters.

[10]  W. Singer,et al.  Temporal coding in the visual cortex: new vistas on integration in the nervous system , 1992, Trends in Neurosciences.

[11]  Yaser S. Abu-Mostafa,et al.  On the K-Winners-Take-All Network , 1988, NIPS.

[12]  M. Fazeli.  Synaptic plasticity: on the trail of the retrograde messenger , 1992, Trends in Neurosciences.

[13]  Roman Bek,et al.  Discourse on one way in which a quantum-mechanics language on the classical logical base can be built up , 1978, Kybernetika.

[14]  S. Kaplan.  The Physiology of Thought , 1950 .

[15]  G. Shepherd.  The Synaptic Organization of the Brain , 1979 .

[16]  Teuvo Kohonen,et al.  Self-organization and associative memory: 3rd edition , 1989 .

[17]  Teuvo Kohonen,et al.  The self-organizing map , 1990 .

[18]  Teuvo Kohonen,et al.  An introduction to neural computing , 1988, Neural Networks.

[19]  B. Gold,et al.  A Comparison of Hamming and Hopfield Neural Nets for Pattern Classification , 1987 .

[20]  Stephen Grossberg,et al.  Nonlinear neural networks: Principles, mechanisms, and architectures , 1988, Neural Networks.

[21]  J. Meador.  Finite precision effects on dynamic behavior in a pulse-coded winner-take-all mechanism , 1992, [Proceedings 1992] IJCNN International Joint Conference on Neural Networks.

[22]  D J Willshaw,et al.  A marker induction mechanism for the establishment of ordered neural mappings: its application to the retinotectal problem. , 1979, Philosophical transactions of the Royal Society of London. Series B, Biological sciences.

[24]  Alan L. Yuille,et al.  A Winner-Take-All Mechanism Based on Presynaptic Inhibition Feedback , 1989, Neural Computation.

[25]  Jui-Chih Yen,et al.  Improved winner-take-all neural network , 1992 .