Hyperfast second-order local solvers for efficient statistically preconditioned distributed optimization

Statistical preconditioning enables fast methods for distributed large-scale empirical risk minimization problems. In this approach, multiple worker nodes compute gradients in parallel, which are then used by the central node to update the parameter by solving an auxiliary (preconditioned) smaller-s...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	EURO journal on computational optimization Jg. 10; S. 100045
Hauptverfasser:	Dvurechensky, Pavel, Kamzolov, Dmitry, Lukashevich, Aleksandr, Lee, Soomin, Ordentlich, Erik, Uribe, César A., Gasnikov, Alexander
Format:	Journal Article
Sprache:	Englisch
Veröffentlicht:	Elsevier Ltd 2022 Elsevier
Schlagworte:	Distributed optimization Empirical risk minimization Statistical preconditioning Tensor optimization methods Empirical risk minimization Distributed optimization Tensor optimization methods Statistical preconditioning
ISSN:	2192-4406
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!