Nearly optimal stochastic approximation for online principal subspace estimation

Principal component analysis (PCA) has been widely used in analyzing high-dimensional data. It converts a set of observed data points of possibly correlated variables into a set of linearly uncorrelated variables via an orthogonal transformation. To handle streaming data and reduce the complexities...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Science China. Mathematics 2023-05, Vol.66 (5), p.1087-1122
Hauptverfasser:	Liang, Xin, Guo, Zhen-Chen, Wang, Li, Li, Ren-Cang, Lin, Wen-Wei
Format:	Artikel
Sprache:	eng
Schlagworte:	Applications of Mathematics Convergence Data points Dimensional analysis Lower bounds Mathematics Mathematics and Statistics Minimax technique Principal components analysis Sampling error Subspaces
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!