Linear Convergence of a Proximal Alternating Minimization Method with Extrapolation for \(\boldsymbol{\ell_1}\) -Norm Principal Component Analysis
SIAM Journal on Optimization · 30.06.2023 · 33 (2) · 684-712
DOI: 10.1137/21m1434507 ↗Kurz gefasst
Fachartikel in der Zeitschrift SIAM Journal on Optimization, online erschienen am 08.06.2023. Verfasst von Peng Wang und 2 weiteren Autor:innen. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 46 Literaturangaben. Gefördert von: ARO YIP, NSF CAREER, Hong Kong RGC GRF project.
Abstract (Originalsprache)
A popular robust alternative of the classic principal component analysis (PCA) is the [Formula: see text]-norm PCA (L1-PCA), which aims to find a subspace that captures the most variation in a dataset as measured by the [Formula: see text]-norm. L1-PCA has shown great promise in alleviating the effect of outliers in data analytic applications. However, it gives rise to a challenging nonsmooth, nonconvex optimization problem, for which existing algorithms are either not scalable or lack strong theoretical guarantees on their convergence behavior. In this paper, we propose a proximal alternating minimization method with extrapolation (PAMe) for solving a two-block reformulation of the L1-PCA problem. We then show that for both the L1-PCA problem and its two-block reformulation, the Kurdyka–Łojasiewicz exponent at any of the limiting critical points is [Formula: see text]. This allows us to establish the linear convergence of the sequence of iterates generated by PAMe and to determine the criticality of the limit of the sequence with respect to both the L1-PCA problem and its two-block reformulation. To complement our theoretical development, we show via numerical experiments on both synthetic and real-world datasets that PAMe is competitive with a host of existing methods. Our results not only significantly advance the convergence theory of iterative methods for L1-PCA but also demonstrate the potential of our proposed method in applications.
Förderung
- ARO YIP
- NSF CAREER
- Hong Kong RGC GRF project
Zitieren
APA
Wang, P., Liu, H., & So, A. M. (2023). Linear Convergence of a Proximal Alternating Minimization Method with Extrapolation for \(\boldsymbol{\ell_1}\) -Norm Principal Component Analysis. SIAM Journal on Optimization, 33(2), 684-712. https://doi.org/10.1137/21m1434507
Vancouver
Wang P, Liu H, So AM. Linear Convergence of a Proximal Alternating Minimization Method with Extrapolation for \(\boldsymbol{\ell_1}\) -Norm Principal Component Analysis. SIAM Journal on Optimization. 2023;33(2):684-712. doi:10.1137/21m1434507
BibTeX
@article{Wang2023,
author = {Wang, Peng and Liu, Huikang and So, Anthony Man-Cho},
title = {Linear Convergence of a Proximal Alternating Minimization Method with Extrapolation for \(\boldsymbol\ell_1\) -Norm Principal Component Analysis},
journal = {SIAM Journal on Optimization},
year = {2023},
volume = {33},
number = {2},
pages = {684--712},
doi = {10.1137/21m1434507}
}
EUVIMED