Spectral Bounds for Kernel Quadrature

DOI10.48550/arXiv.2609.00553MaRDI QIDQ7285127arXivFDO

Authors Alexander Cloninger, Q. T. Le Gia, H. N. Mhaskar, A. Cloninger

Publication date 1 September 2026

Published in arXiv

Tldr We introduce novel spectral bounds for approximating Gaussian process regression via kernel quadrature with explicit feature maps, improving the accuracy of existing bounds by up to 40% and reducing the computational cost by up to 90%.

Full work available at URL https://doi.org/10.48550/arXiv.2609.00553

arXiv classification

math.NA

This page was built for publication: Spectral Bounds for Kernel Quadrature

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7285127)