Work

Parameter estimation in the Hermitian and skew‐Hermitian splitting method using gradient iterations

Year: 2020

Type: article

Abstract: Summary This article presents enhancement strategies for the Hermitian and skew‐Hermitian splitting method based on gradient iterations. The spectral properties are exploited for the parameter estimation, often resulting in a better convergence. In particular, steepest descent with early stopping can generate a rough estimate of the optimal upper bound. This is better than an arbitrary choice since the latter often causes stability problems or slow convergence. In addition, delayed gradient methods are considered as inner solvers for the splitting method. Experiments verify the effectiveness of the proposed estimation strategies and show that delayed gradient methods are competitive with conjugate gradient in low precision.

Source: Numerical Linear Algebra with Applications

DOI: 10.1002/nla.2304

Authors: Qinmeng Zou, Frédéric Magoulès

Institutions: Beijing University of Posts and Telecommunications, Université Paris-Saclay, CentraleSupélec, University of Pecs

Language: English

Topic: Matrix Theory and Algorithms

Subfield: Computational Theory and Mathematics

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Estimated APC paid: $3,600

Funders: Agence Nationale de la Recherche, Centre National de la Recherche Scientifique