Randomized symmetric Gauss-Seidel method for solving linear least squares problems

被引:0
作者
Sha, Fan [1 ]
Zhang, Jianbing [2 ]
机构
[1] East China Normal Univ, Sch Math, Shanghai 200241, Peoples R China
[2] Nanjing Univ, Sch Artificial Intelligence, Nanjing 210023, Peoples R China
来源
AIMS MATHEMATICS | 2024年 / 9卷 / 07期
基金
国家重点研发计划;
关键词
randomized symmetric Gauss-Seidel; linear least squares; randomized sampling; project; converge in expectation; probability criterion; KACZMARZ;
D O I
10.3934/math.2024848
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
We introduced a random symmetric Gauss-Seidel (RSGS) method, which was designed to handle large scale linear least squares problems involving tall coefficient matrices. This RSGS method projected the approximate residual onto the subspace spanned by two symmetric columns at each iteration. These columns were sampled from the coefficient matrix based on an effective probability criterion. Our theoretical analysis indicated that RSGS converged when the coefficient matrix had full column rank. Furthermore, numerical experiments demonstrated that RSGS outperformed the baseline algorithms in terms of iteration steps and CPU time.
引用
收藏
页码:17453 / 17463
页数:11
相关论文
共 16 条
[1]   A two-step randomized Gauss-Seidel method for solving large-scale linear least squares problems [J].
不详 .
ELECTRONIC RESEARCH ARCHIVE, 2022, 30 (02) :755-779
[2]   On greedy randomized coordinate descent methods for solving large linear least-squares problems [J].
Bai, Zhong-Zhi ;
Wu, Wen-Ting .
NUMERICAL LINEAR ALGEBRA WITH APPLICATIONS, 2019, 26 (04)
[3]   The University of Florida Sparse Matrix Collection [J].
Davis, Timothy A. ;
Hu, Yifan .
ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE, 2011, 38 (01)
[4]   Tight upper bounds for the convergence of the randomized extended Kaczmarz and Gauss-Seidel algorithms [J].
Du, Kui .
NUMERICAL LINEAR ALGEBRA WITH APPLICATIONS, 2019, 26 (03)
[5]  
Golub G. H., 2013, Matrix Computations, V4th ed.
[6]   RANDOMIZED ITERATIVE METHODS FOR LINEAR SYSTEMS [J].
Gower, Robert M. ;
Richtarik, Peter .
SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 2015, 36 (04) :1660-1690
[7]   A Note On Convergence Rate of Randomized Kaczmarz Method [J].
Guan, Ying-Jun ;
Li, Wei-Guo ;
Xing, Li-Li ;
Qiao, Tian-Tian .
CALCOLO, 2020, 57 (03)
[8]   ROWS VERSUS COLUMNS: RANDOMIZED KACZMARZ OR GAUSS-SEIDEL FOR RIDGE REGRESSION [J].
Hefny, Ahmed ;
Needell, Deanna ;
Ramdas, Aaditya .
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2017, 39 (05) :S528-S542
[9]   Randomized Methods for Linear Constraints: Convergence Rates and Conditioning [J].
Leventhal, D. ;
Lewis, A. S. .
MATHEMATICS OF OPERATIONS RESEARCH, 2010, 35 (03) :641-654
[10]   On maximum residual block and two-step Gauss-Seidel algorithms for linear least-squares problems [J].
Liu, Yong ;
Jiang, Xiang-Long ;
Gu, Chuan-Qing .
CALCOLO, 2021, 58 (02)