ヒダカ　アキノリ   HIDAKA Akinori   日高　章理    所属   東京電機大学  理工学部 理工学科　理学系    東京電機大学大学院  理工学研究科 理学専攻    東京電機大学大学院  先端科学技術研究科 数理学専攻    職種   准教授
言語種別	英語
発行・発表の年月	2018/01
形態種別	学術研究論文
査読	査読有り
標題	Data Visualization for Deep Neural Networks Based on Interlayer Canonical Correlation Analysis
執筆形態	共著
掲載誌名	Transactions of ISCIE
掲載区分	国内
出版社・発行元	The Institute of Systems, Control and Information Engineers (ISCIE)
巻・号・頁	31(1),pp.10-20
著者・共著者	◎Akinori Hidaka and Takio Kurita
概要	In this paper, we develop data visualization methods which consider interlayer correlations in deep neural networks (DNN). In general, DNN naturally acquires multiple feature representations corresponding to their intermediate layers through their learning process. In order to understand relationships of those intermediate features which are strongly correlated with each other, we utilize canonical correlation analysis (CCA) to visualize the data distributions of different feature layers in a common subspace. Our method can grasp movement of samples between consecutive layers in DNN. By using standard benchmark data sets, we show that our visualization results contain much information that typical visualization methods (such as principal component analysis) usually do not represent.