Parametric Kernel Low-Rank Approximations Using Tensor Train Decomposition
SIAM Journal on Matrix Analysis and Applications · 30.06.2025 · 46 (2) · 1006-1036
DOI: 10.1137/24m1663879 ↗Kurz gefasst
Fachartikel in der Zeitschrift SIAM Journal on Matrix Analysis and Applications, online erschienen am 09.04.2025. Verfasst von Abraham Khan und 1 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 45 Literaturangaben.
Abstract (Originalsprache)
Computing low-rank approximations of kernel matrices is an important problem with many applications in scientific computing and data science. We propose methods to efficiently approximate and store low-rank approximations to kernel matrices that depend on certain hyperparameters. The main idea behind our method is to use multivariate Chebyshev function approximation along with the tensor train decomposition of the coefficient tensor. The computations are in two stages: an offline stage, which dominates the computational cost and is parameter-independent, and an online stage, which is inexpensive and instantiated for specific hyperparameters. A variation of this method addresses the case that the kernel matrix is symmetric and positive semidefinite. The resulting algorithms have linear complexity in terms of the sizes of the kernel matrices. We investigate the efficiency and accuracy of our method on parametric kernel matrices induced by various kernels, such as the Matérn kernel, through various numerical experiments on problem setups up to three spatial dimensions and two parameter dimensions. Our methods have speedups up to [Formula: see text] in the online time compared to other methods with similar complexity and comparable accuracy. Reproducibility of computational results. This paper has been awarded the “SIAM Reproducibility Badge: Code and data available” as a recognition that the authors have followed reproducibility principles valued by SIMAX and the scientific computing community. Code and data that allow readers to reproduce the results in this paper are available at https://github.com/awkhan3/ParametricTensorTrainKernel and in the supplemental material ( Supp_Materials.zip [3.60MB]). [Formula: see text]
Zitieren
APA
Khan, A., & Saibaba, A. K. (2025). Parametric Kernel Low-Rank Approximations Using Tensor Train Decomposition. SIAM Journal on Matrix Analysis and Applications, 46(2), 1006-1036. https://doi.org/10.1137/24m1663879
Vancouver
Khan A, Saibaba AK. Parametric Kernel Low-Rank Approximations Using Tensor Train Decomposition. SIAM Journal on Matrix Analysis and Applications. 2025;46(2):1006-1036. doi:10.1137/24m1663879
BibTeX
@article{Khan2025,
author = {Khan, Abraham and Saibaba, Arvind K.},
title = {Parametric Kernel Low-Rank Approximations Using Tensor Train Decomposition},
journal = {SIAM Journal on Matrix Analysis and Applications},
year = {2025},
volume = {46},
number = {2},
pages = {1006--1036},
doi = {10.1137/24m1663879}
}
EUVIMED