Deflated HeteroPCA: Overcoming the curse of ill-conditioning in heteroskedastic PCA
The Annals of Statistics · 01.02.2025 · 53 (1) · 91-116
DOI: 10.1214/24-aos2456 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Annals of Statistics, erschienen am 01.02.2025. Verfasst von Yuchen Zhou und 1 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 82 Literaturangaben.
Abstract (Originalsprache)
This paper is concerned with estimating the column subspace of a low-rank matrix X⋆∈Rn1×n2 from contaminated data. How to obtain optimal statistical accuracy while accommodating the widest range of signal-to-noise ratios (SNRs) becomes particularly challenging in the presence of heteroskedastic noise and unbalanced dimensionality (i.e., n2≫n1). While the state-of-the-art algorithm HeteroPCA emerges as a powerful solution for solving this problem, it suffers from “the curse of ill-conditioning,” namely, its performance degrades as the condition number of X⋆ grows. In order to overcome this critical issue without compromising the range of allowable SNRs, we propose a novel algorithm, called Deflated−HeteroPCA, that achieves near-optimal and condition-number-free theoretical guarantees in terms of both ℓ2 and ℓ2,∞ statistical accuracy. The proposed algorithm divides the spectrum of X⋆ into well-conditioned and mutually well-separated subblocks, and applies HeteroPCA to conquer each subblock successively. Further, an application of our algorithm and theory to two canonical examples—the factor model and tensor PCA—leads to remarkable improvement for each application.
Zitieren
APA
Zhou, Y., & Chen, Y. (2025). Deflated HeteroPCA: Overcoming the curse of ill-conditioning in heteroskedastic PCA. The Annals of Statistics, 53(1), 91-116. https://doi.org/10.1214/24-aos2456
Vancouver
Zhou Y, Chen Y. Deflated HeteroPCA: Overcoming the curse of ill-conditioning in heteroskedastic PCA. The Annals of Statistics. 2025;53(1):91-116. doi:10.1214/24-aos2456
BibTeX
@article{Zhou2025,
author = {Zhou, Yuchen and Chen, Yuxin},
title = {Deflated HeteroPCA: Overcoming the curse of ill-conditioning in heteroskedastic PCA},
journal = {The Annals of Statistics},
year = {2025},
volume = {53},
number = {1},
pages = {91--116},
doi = {10.1214/24-aos2456}
}
EUVIMED