Abstract
Although no distance function over the input data is definable, it is still possible to implement the self-organizing map (SOM) process using evolutionary-learning operations. The process can be made to converge more rapidly when the probabilistic trials of conventional evolutionary learning are replaced by averaging using the so-called Batch Map version of the self-organizing map. Although no other condition or metric than a fitness function between the input samples and the models is assumed, an order in the map that complies with the ‘functional similarity’ of the models can be seen to emerge. There exist two modes of use of this new principle: representation of nonmetric input data distributions by models that may have variable structures, and fast generation of evolutionary cycles that resemble those defined by the genetic algorithms. The spatial order in the array of models can be utilized for finding more uniform variations, such as crossings between functionally similar models.
Similar content being viewed by others
References
T. Kohonen, Self-Organization and Associative Memory, Springer-Verlag, Berlin, 1985, 3rd edition, 1989.
T. Kohonen, Self-Organizing Maps, Springer-Verlag, Berlin, 1995, 2nd edition, 1997.
D. Polani and T. Uthmann, “Adaptation of Kohonen feature map topologies by genetic algorithms”, in R. Männer and B. Manderick (eds), Parallel Problem Solving from Nature, Elsevier, pp. 421–429, 1992.
H. Hyötyniemi, A.S. Nissinen, and H.N. Koivo, “Evolution based self-organization of structures in linear time-series modeling”, Proc. of the Third Nordic Workshop on Genetic Algorithms and their Applications (3NWGA), Helsinki, Finland, August 20-22, pp. 135–152.
A.S. Nissinen and H. Hyötyniemi, “Evolutionary training of behavior-based self-organizing map”, Proc. of the World Congress on Computational Intelligence, WCCI'98, Alaska, May 1998, to appear.
T. Kohonen, “New developments of learning vector quantization and the self-organizing map”, SYNAPSE'92, Symposium on Neural Networks; Alliances and Perspectives in Senri 1992, Osaka, Japan, June 24-26, 1992.
J. Kangas and S. Kaski, “3043 works that have been based on the self-organizing map (SOM) method developed by Kohonen”, Report A49, Helsinki University of Technology, Laboratory of Computer and Information Science, Espoo, Finland, 1998. Also available in Internet at the address http://www.cis.hut.fi/nnrc/refs/.
Y. Cheng, “Convergence and ordering of Kohonen's batch map”, Neural Computation, Vol. 9, pp. 1667–1676, 1997.
J. Makhoul, S. Roucos, and H. Gish, “Vector quantization in speech coding”, Proc. IEEE, Vol. 73, pp. 1551–1588, 1985.
H. Ritter and K. Schulten, “On the stationary state of Kohonen's self-organizing sensory mapping”, Biol. Cyb., Vol. 54, pp. 99–106, 1986.
H. Ritter and K. Schulten, “Convergence properties of Kohonen's topology preserving maps: Fluctuations, stability, and dimension selection”, Biol. Cyb., Vol. 60, pp. 59–71, 1988.
M. Herrmann, H.-U. Bauer, and R. Der, “Optimal magnification factors in self-organizing feature maps”, in F. Fogelman-Soulié and P. Gallinari (eds) Proc. ICANN'95, International Conference on Artificial Neural Networks, Vol. 1, pp. 75–80, EC2, Nanterre, France, 1995.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Kohonen, T. Fast Evolutionary Learning with Batch-Type Self-Organizing Maps. Neural Processing Letters 9, 153–162 (1999). https://doi.org/10.1023/A:1018681526204
Issue Date:
DOI: https://doi.org/10.1023/A:1018681526204