Title :
Generalizations of the self-organizing map
Author_Institution :
Lab. of Comput. & Inf. Sci., Helsinki Univ. of Technol., Espoo, Finland
Abstract :
It is pointed out in this paper that for the cells of a self-organizing map (SOM) one may select adaptive dynamic operators in stead of simple static units. Their updating may be made either by tuning their parameters, or in an evolutionary process. This kind of generic principle is supposed to define the category of SOM networks in a more general sense.
Keywords :
self-organising feature maps; adaptive dynamic operators; evolutionary process; parameter tuning; self-organizing map; updating; Automatic control; Control systems; Filters; Information science; Laboratories; Nervous system; Neurons; Output feedback; Polynomials; Signal processing;
Conference_Titel :
Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
Print_ISBN :
0-7803-1421-2
DOI :
10.1109/IJCNN.1993.713953