Online Learning as Stochastic Approximation of Regularization Paths: Optimality and Almost-Sure Convergence

In this paper, an online learning algorithm is proposed as sequential stochastic approximation of a regularization path converging to the regression function in reproducing kernel Hilbert spaces (RKHSs). We show that it is possible to produce the best known strong (RKHS norm) convergence rate of bat...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	IEEE transactions on information theory 2014-09, Vol.60 (9), p.5716-5735
Hauptverfasser:	Tarres, Pierre, Yuan Yao
Format:	Artikel
Sprache:	eng
Schlagworte:	Applied sciences Approximation Approximation methods Convergence Educational institutions Exact sciences and technology Hilbert space Information theory Information, signal and communications theory Kernel Probabilistic logic Stochastic models Telecommunications and information theory Upper bound
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!