Abstract
In this article a convergence criterion for neural networks of competitive learning alternative to that established by Rumelhart et al. [Rumel86] is presented. With it the number of iterations needed so that the network reaches a stable configuration is reduced to a high degree. The new convergence criterion is based on the network stability measurement in contrast to the weight variation which is defined by Rumelhart et al. Results obtained in a number of realized tests which allow you to evaluate the step number reduction reached when applying the new criterion as contrasted with Rumelharts are shown.
This is a preview of subscription content, log in via an institution.
Preview
Unable to display preview. Download preview PDF.
Bibliography
Rumelhart D. y McClelland J.."Parallel Distributed Processing. Exploration in the Microstructure of cognition". Vol.1 1986.
Sacristán A.. "Criterio de convergencia para redes con aprendizaje competitivo". Trabajo presentado como Tesina. 1990
Rumelhart D. y Zipser D. "Feature discovery by Competitive Learning". Cognitive Science. 1985. Pag.75–112
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1991 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Sacristán, A., Valderrama, E., Pérez-Vicente, C. (1991). Stability measurement criterion for neural networks of competitive learning. In: Prieto, A. (eds) Artificial Neural Networks. IWANN 1991. Lecture Notes in Computer Science, vol 540. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0035879
Download citation
DOI: https://doi.org/10.1007/BFb0035879
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-54537-8
Online ISBN: 978-3-540-38460-1
eBook Packages: Springer Book Archive