A Fast Proximal Alternating Method for Robust Matrix Factorization of Matrix Recovery with Outliers

This paper concerns a class of robust factorization models of low-rank matrix recovery, which have been widely applied in various fields such as machine learning and imaging sciences. An <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline">...

Full description

Saved in:
Bibliographic Details
Main Authors: Ting Tao, Lianghai Xiao, Jiayuan Zhong
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Mathematics
Subjects:
Online Access:https://www.mdpi.com/2227-7390/13/9/1466
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850032202655465472
author Ting Tao
Lianghai Xiao
Jiayuan Zhong
author_facet Ting Tao
Lianghai Xiao
Jiayuan Zhong
author_sort Ting Tao
collection DOAJ
description This paper concerns a class of robust factorization models of low-rank matrix recovery, which have been widely applied in various fields such as machine learning and imaging sciences. An <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mo>ℓ</mo><mn>1</mn></msub></semantics></math></inline-formula>-loss robust factorized model incorporating the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mo>ℓ</mo><mrow><mn>2</mn><mo>,</mo><mn>0</mn></mrow></msub></semantics></math></inline-formula>-norm regularization term is proposed to address the presence of outliers. Since the resulting problem is nonconvex, nonsmooth, and discontinuous, an approximation problem that shares the same set of stationary points as the original formulation is constructed. Subsequently, a proximal alternating minimization method is proposed to solve the approximation problem. The global convergence of its iterate sequence is also established. Numerical experiments on matrix completion with outliers and image restoration tasks demonstrate that the proposed algorithm achieves low relative errors in shorter computational time, especially for large-scale datasets.
format Article
id doaj-art-88e9ea177bb74a32ac1c53b1986a88e2
institution DOAJ
issn 2227-7390
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Mathematics
spelling doaj-art-88e9ea177bb74a32ac1c53b1986a88e22025-08-20T02:58:44ZengMDPI AGMathematics2227-73902025-04-01139146610.3390/math13091466A Fast Proximal Alternating Method for Robust Matrix Factorization of Matrix Recovery with OutliersTing Tao0Lianghai Xiao1Jiayuan Zhong2School of Mathematics, Foshan University, Foshan 528011, ChinaCollege of Information Science and Technology, Jinan University, Guangzhou 510632, ChinaSchool of Mathematics, Foshan University, Foshan 528011, ChinaThis paper concerns a class of robust factorization models of low-rank matrix recovery, which have been widely applied in various fields such as machine learning and imaging sciences. An <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mo>ℓ</mo><mn>1</mn></msub></semantics></math></inline-formula>-loss robust factorized model incorporating the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mo>ℓ</mo><mrow><mn>2</mn><mo>,</mo><mn>0</mn></mrow></msub></semantics></math></inline-formula>-norm regularization term is proposed to address the presence of outliers. Since the resulting problem is nonconvex, nonsmooth, and discontinuous, an approximation problem that shares the same set of stationary points as the original formulation is constructed. Subsequently, a proximal alternating minimization method is proposed to solve the approximation problem. The global convergence of its iterate sequence is also established. Numerical experiments on matrix completion with outliers and image restoration tasks demonstrate that the proposed algorithm achieves low relative errors in shorter computational time, especially for large-scale datasets.https://www.mdpi.com/2227-7390/13/9/1466matrix recovery with outliersglobal convergencecolumn ℓ2,0-normalternating method
spellingShingle Ting Tao
Lianghai Xiao
Jiayuan Zhong
A Fast Proximal Alternating Method for Robust Matrix Factorization of Matrix Recovery with Outliers
Mathematics
matrix recovery with outliers
global convergence
column ℓ2,0-norm
alternating method
title A Fast Proximal Alternating Method for Robust Matrix Factorization of Matrix Recovery with Outliers
title_full A Fast Proximal Alternating Method for Robust Matrix Factorization of Matrix Recovery with Outliers
title_fullStr A Fast Proximal Alternating Method for Robust Matrix Factorization of Matrix Recovery with Outliers
title_full_unstemmed A Fast Proximal Alternating Method for Robust Matrix Factorization of Matrix Recovery with Outliers
title_short A Fast Proximal Alternating Method for Robust Matrix Factorization of Matrix Recovery with Outliers
title_sort fast proximal alternating method for robust matrix factorization of matrix recovery with outliers
topic matrix recovery with outliers
global convergence
column ℓ2,0-norm
alternating method
url https://www.mdpi.com/2227-7390/13/9/1466
work_keys_str_mv AT tingtao afastproximalalternatingmethodforrobustmatrixfactorizationofmatrixrecoverywithoutliers
AT lianghaixiao afastproximalalternatingmethodforrobustmatrixfactorizationofmatrixrecoverywithoutliers
AT jiayuanzhong afastproximalalternatingmethodforrobustmatrixfactorizationofmatrixrecoverywithoutliers
AT tingtao fastproximalalternatingmethodforrobustmatrixfactorizationofmatrixrecoverywithoutliers
AT lianghaixiao fastproximalalternatingmethodforrobustmatrixfactorizationofmatrixrecoverywithoutliers
AT jiayuanzhong fastproximalalternatingmethodforrobustmatrixfactorizationofmatrixrecoverywithoutliers