Matrices With Gaussian Noise: Optimal Estimates for Singular Subspace Perturbation
IEEE Transactions on Information Theory · 2024-03 · 70 (3) · 1978-2002
DOI: 10.1109/tit.2023.3331010 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Information Theory, 2024-03. Verfasst von Sean O’Rourke und 2 weiteren Autor:innen. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 77 Literaturangaben. Gefördert von: National Science Foundation, Hong Kong Research Grants Council.
Abstract (Originalsprache)
The Davis–Kahan–Wedin $\sin \Theta $ theorem describes how the singular subspaces of a matrix change when subjected to a small perturbation. This classic result is sharp in the worst case scenario. In this paper, we prove a stochastic version of the Davis–Kahan–Wedin $\sin \Theta $ theorem when the perturbation is a Gaussian random matrix. Under certain structural assumptions, we obtain an optimal bound that significantly improves upon the classic Davis–Kahan–Wedin $\sin \Theta $ theorem. One of our key tools is a new perturbation bound for the singular values, which may be of independent interest.
Förderung
- National Science Foundation
- Hong Kong Research Grants Council
Zitieren
APA
O’Rourke, S., Vu, V., & Wang, K. (2024). Matrices With Gaussian Noise: Optimal Estimates for Singular Subspace Perturbation. IEEE Transactions on Information Theory, 70(3), 1978-2002. https://doi.org/10.1109/tit.2023.3331010
Vancouver
O’Rourke S, Vu V, Wang K. Matrices With Gaussian Noise: Optimal Estimates for Singular Subspace Perturbation. IEEE Transactions on Information Theory. 2024;70(3):1978-2002. doi:10.1109/tit.2023.3331010
BibTeX
@article{ORourke2024,
author = {O’Rourke, Sean and Vu, Van and Wang, Ke},
title = {Matrices With Gaussian Noise: Optimal Estimates for Singular Subspace Perturbation},
journal = {IEEE Transactions on Information Theory},
year = {2024},
volume = {70},
number = {3},
pages = {1978--2002},
doi = {10.1109/tit.2023.3331010}
}
EUVIMED