A Topology Preserving Neural Network for Nonstationary Distributions

Taira NAKAJIMA  Hiroyuki TAKIZAWA  Hiroaki KOBAYASHI  Tadao NAKAMURA  

Publication
IEICE TRANSACTIONS on Information and Systems   Vol.E82-D   No.7   pp.1131-1135
Publication Date: 1999/07/25
Online ISSN: 
DOI: 
Print ISSN: 0916-8532
Type of Manuscript: LETTER
Category: Bio-Cybernetics and Neurocomputing
Keyword: 
competitive Hebbian learning rule,  law of the jungle mechanism,  neural network,  nonstationary probability distribution,  self-organizing map,  

Full Text: PDF(209.4KB)>>
Buy this Article




Summary: 
We propose a learning algorithm for self-organizing neural networks to form a topology preserving map from an input manifold whose topology may dynamically change. Experimental results show that the network using the proposed algorithm can rapidly adjust itself to represent the topology of nonstationary input distributions.