Summary
International Symposium on Nonlinear Theory and its Applications
2008
Session Number:B3L-D
Session:
Number:B3L-D2
A Cellular Structual Analysis for Covariance Structure
Ryo Sekiyama, Hisashi Aomori, Mamoru Tanaka,
pp.-
Publication Date:2008/9/7
Online ISSN:2188-5079
Summary:
The cellular computing concept is very powerful method for a locally connected network such like cellular neural network. Owing to its parallelization performance, this method is very suitable for a large scale problem. The covariance structure analysis is one of the well-known method for validating the given model. However, its computational cost is very high. In this paper, we propose a novel covariance structure analysis method based on cellular neural network for fast processing. Moreover, this method enables the CNN to acquire learning ability. The experiment results suggest that the proposed method is an equivalence frame work of the conventional covariance structure analysis.