A Sketched Generalized Krylov Subspace Method for Large-Scale Regularization

DOI10.48550/arXiv.2606.18073MaRDI QIDQ7267620zbMATH OpenarXivFDO

Authors Davide Palitta, Mirjeta Pasha

Publication date 16 June 2026

Published in arXiv

Tldr This paper presents sGKS, a sketched generalized Krylov subspace method eliminating computational bottlenecks in large-scale Tikhonov regularization by compressing matrices and skipping reorthogonalization without compromising approximation quality. Results demonstrate this approach matches standard reconstruction accuracy while significantly reducing per-iteration costs through incremental updates with optional iterative refinement for challenging scenarios.

Full work available at URL https://doi.org/10.48550/arXiv.2606.18073

arXiv classification

math.NA

This page was built for publication: A Sketched Generalized Krylov Subspace Method for Large-Scale Regularization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7267620)