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

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...

Full description

Bibliographic Details
Main Authors: Pavel Dvurechensky, Dmitry Kamzolov, Aleksandr Lukashevich, Soomin Lee, Erik Ordentlich, César A. Uribe, Alexander Gasnikov
Format: Article
Language:English
Published: Elsevier 2022-01-01
Series:EURO Journal on Computational Optimization
Subjects:
Empirical risk minimization
Distributed optimization
Statistical preconditioning
Tensor optimization methods
Online Access:http://www.sciencedirect.com/science/article/pii/S2192440622000211

Internet

http://www.sciencedirect.com/science/article/pii/S2192440622000211

Similar Items