広島大学大学院工学研究科研究報告 53 巻 1 号
2004-12 発行

誤差逆伝播学習法による自己組織化ロバスト主成分分析

Self-Organized Robust Principal Component Analysis by Back-Propagation Learning
樋口 勇夫
全文
274 KB
KJ00004291002.pdf
Abstract
The purpose of this study is the suggestion of a self-organized back-propagation algorithm for robust principal component analysis. The self-organizing algorithm that discriminates the influence of data automatically is applied to learning of a sandglass type neural network.
著者キーワード
robust principal component analysis
neural networks
back-propagation
self-organizing rule