On greedy randomized coordinate descent methods for solving large linear least-squares problems

被引:77
作者
Bai, Zhong-Zhi [1 ,2 ]
Wu, Wen-Ting [1 ,2 ]
机构
[1] Chinese Acad Sci, Acad Math & Syst Sci, State Key Lab Sci Engn Comp, Inst Computat Math & Sci Engn Comp, POB 2719, Beijing 100190, Peoples R China
[2] Univ Chinese Acad Sci, Sch Math Sci, Beijing, Peoples R China
基金
中国国家自然科学基金;
关键词
convergence property; coordinate descent method; linear least-squares problem; randomized iteration; CONVERGENCE; OPTIMIZATION; ALGORITHMS; TOMOGRAPHY; EFFICIENCY;
D O I
10.1002/nla.2237
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
For solving large scale linear least-squares problem by iteration methods, we introduce an effective probability criterion for selecting the working columns from the coefficient matrix and construct a greedy randomized coordinate descent method. It is proved that this method converges to the unique solution of the linear least-squares problem when its coefficient matrix is of full rank, with the number of rows being no less than the number of columns. Numerical results show that the greedy randomized coordinate descent method is more efficient than the randomized coordinate descent method.
引用
收藏
页数:15
相关论文
共 31 条
[1]   ON GREEDY RANDOMIZED KACZMARZ METHOD FOR SOLVING LARGE SPARSE LINEAR SYSTEMS [J].
Bai, Zhong-Zhi ;
Wu, Wen-Ting .
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2018, 40 (01) :A592-A606
[2]  
Bjorck A., 1996, Numerical Methods for Least Squares Problems, DOI DOI 10.1137/1.9781611971484
[3]   A unified approach to statistical tomography using coordinate descent optimization [J].
Bouman, CA ;
Sauer, K .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 1996, 5 (03) :480-492
[4]   COORDINATE DESCENT ALGORITHMS FOR NONCONVEX PENALIZED REGRESSION, WITH APPLICATIONS TO BIOLOGICAL FEATURE SELECTION [J].
Breheny, Patrick ;
Huang, Jian .
ANNALS OF APPLIED STATISTICS, 2011, 5 (01) :232-253
[5]   Cyclic coordinate descent: A robotics algorithm for protein loop closure [J].
Canutescu, AA ;
Dunbrack, RL .
PROTEIN SCIENCE, 2003, 12 (05) :963-972
[6]   OPTIMIZATION OF LOG-X ENTROPY OVER LINEAR EQUALITY CONSTRAINTS [J].
CENSOR, Y ;
LENT, A .
SIAM JOURNAL ON CONTROL AND OPTIMIZATION, 1987, 25 (04) :921-933
[7]  
Chang KW, 2008, J MACH LEARN RES, V9, P1369
[8]   The University of Florida Sparse Matrix Collection [J].
Davis, Timothy A. ;
Hu, Yifan .
ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE, 2011, 38 (01)
[9]  
DEMMEL JW, 1988, MATH COMPUT, V50, P449, DOI 10.1090/S0025-5718-1988-0929546-7
[10]  
Golub G.H., 2013, Matrix Computations, V4th, DOI [10.56021/9781421407944, DOI 10.56021/9781421407944]