On stochastic gradient and subgradient methods with adaptive steplength sequences

Traditionally, stochastic approximation (SA) schemes have been popular choices for solving stochastic optimization problems. However, the performance of standard SA implementations can vary significantly based on the choice of the steplength sequence, and in general, little guidance is provided abou...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	Automatica (Oxford) Jg. 48; H. 1; S. 56 - 67
Hauptverfasser:	Yousefian, Farzad, Nedić, Angelia, Shanbhag, Uday V.
Format:	Journal Article
Sprache:	Englisch
Veröffentlicht:	Kidlington Elsevier Ltd 2012 Elsevier
Schlagworte:	Adaptive steplength Applied sciences Approximation Convergence Convex optimization Decision theory. Utility theory Exact sciences and technology Mathematical analysis Mathematical programming Operational research and scientific management Operational research. Management science Optimization Permissible error Randomized smoothing techniques Sampling Stochastic approximation Stochastic optimization Stochasticity Stochastic approximation Randomized smoothing techniques Stochastic optimization Convex optimization Adaptive steplength Perturbation function Lipschitz condition Approximation algorithm Function approximation Stochastic method Adaptive method Convex programming Uniform distribution Randomization Lipschitz function Sampling Gradient method Subgradient optimization Error bound Statistical decision Standards Stochastic programming Guidance Smoothing Public key Randomness Recursive method Problem solving Objective function
ISSN:	0005-1098, 1873-2836
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Schreiben Sie den ersten Kommentar!