Direct optimization of BPX preconditioners

Ivan Oseledets, Vladimir Fanaskov

    Research output: Contribution to journalArticlepeer-review

    Abstract

    We consider an automatic construction of locally optimal preconditioners for positive definite linear systems. To achieve this goal, we introduce a differentiable loss function that does not explicitly include the estimation of minimal eigenvalue. Nevertheless, the resulting optimization problem is equivalent to a direct minimization of the condition number. To demonstrate our approach, we construct a parametric family of modified BPX preconditioners. Namely, we define a set of empirical basis functions for coarse finite element spaces and tune them to achieve better condition number. For considered model equations (that includes Poisson, Helmholtz, Convection–diffusion, Biharmonic, and others), we achieve from two to twenty times smaller condition numbers for symmetric positive definite linear systems.

    Original languageEnglish
    Article number113811
    JournalJournal of Computational and Applied Mathematics
    Volume402
    DOIs
    Publication statusPublished - 1 Mar 2022

    Keywords

    • Boundary value problems
    • BPX
    • Multigrid
    • Numerical linear algebra
    • PDE
    • Preconditioning

    Fingerprint

    Dive into the research topics of 'Direct optimization of BPX preconditioners'. Together they form a unique fingerprint.

    Cite this