Summary
International Symposium on Nonlinear Theory and its Applications
2008
Session Number:C3L-E
Session:
Number:C3L-E1
Lazy Self-Organizing Map Considering Lazy-Neuron Rate for Effective Self-Organization
Taku Haraguchi, Haruna Matsushita, Yoshifumi Nishio ,
pp.-
Publication Date:2008/9/7
Online ISSN:2188-5079
DOI:10.34385/proc.42.C3L-E1
PDF download (1MB)
Summary:
In the previous study, we have proposed the Lazy Self-Organizing Map (LSOM), whose learning rate depends on only each neuron’s character, and investigated its behaviors. In this study, we propose an improved LSOM, whose learning rate depends on each neuron’s character, lazy-neuron rate and time. We apply the improved LSOM considering lazy-neuron rate to various input data set and investigate its effectiveness using three measurements.