Abstract
We investigate how to use an LU factorization with the classical lsqr routine for solving overdetermined sparse least squares problems. Usually L is much better conditioned than A and iterating with L instead of A results in faster convergence. When a runtime test indicates that L is not sufficiently well-conditioned, a partial orthogonalization of L accelerates the convergence. Numerical experiments illustrate the good behavior of our algorithm in terms of storage and convergence.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Amestoy, P.R., Guermouche, A., L’Excellent, J.-Y., Pralet, S.: MUMPS: a MUltifrontal Parallel sparse direct Solver (2014). http://mumps.enseeiht.fr/index.php?page=home
Baboulin, M., Giraud, L., Gratton, S., Langou, J.: Parallel tools for solving incremental dense least squares problems. Application to space geodesy. J. Algorithms Comput. Technol. 3(1), 117–133 (2009)
Baboulin, M., Gratton, S., Lacroix, R., Laub, A.J.: Statistical estimates for the conditioning of linear least squares problems. In: Wyrzykowski, R., Dongarra, J., Karczewski, K., Waśniewski, J. (eds.) PPAM 2013, Part I. LNCS, vol. 8384, pp. 124–133. Springer, Heidelberg (2014)
Bateman, D., Adler, A.: Sparse matrix implementation in octave (2006). arxiv.org/pdf/cs/0604006.pdf
Benzi, M., Tuma, M.: A robust incomplete factorization preconditioner for positive definite matrices. Numer. Linear Algebra Appl. 10, 385–400 (2003)
Björck, Å.: Numerical Methods for Least Squares Problems. SIAM, Philadelphia (1996)
Björck, A., Duff, I.S.: A direct method for the solution of sparse linear least squares problems. Linear Algebra Appl. 34, 43–67 (1980)
Björck, A., Yuan, J.Y.: Preconditioners for least squares problems by \(LU\) factorization. Electron. Trans. Numer. Anal. 8, 26–35 (1999)
Davis, T.: The University of Florida sparse matrix collection. http://www.cise.ufl.edu/research/sparse/matrices/
Davis, T.: Direct Methods for Sparse Linear Systems. SIAM, Philadelphia (2006)
Golub, G.H., Van Loan, C.F.: Matrix Computations, 3rd edn. The Johns Hopkins University Press, Baltimore (1996)
Gupta, A.: Improved symbolic and numerical factorization algorithms for unsymmetric sparse matrices. SIAM J. Matrix Anal. Appl. 24, 529–552 (2002)
Hager, W.W.: Condition estimates. SIAM J. Sci. Stat. Comput. 5, 311–316 (1984)
Heath, M.T.: Numerical methods for large sparse linear squares problems. SIAM J. Sci. Stat. Comput. 5(4), 497–513 (1984)
Higham, N.J., Tisseur, F.: A block algorithm for matrix 1-norm estimation with an application to 1-norm pseudospectra. SIAM J. Matrix Anal. Appl. 21, 1185–1201 (2000)
Howell, G., Baboulin, M.: Data and plots for lsqr, lsqrLU and lsqrLUQR (2015). http://ncsu.edu/hpc/Documets/Publications/gary_howell/contents.html
Jennings, A., Ajiz, M.A.: Incomplete methods for solving \(A^TAx = b\). SIAM J. Sci. Stat. Comput. 5(4), 978–987 (1984)
Li, N., Saad, Y.: MIQR: a multilevel incomplete \(QR\) preconditioner for large sparse least-squares problems. SIAM J. Matrix Anal. Appl. 28(2), 524–550 (2006)
Li, X.S.: An overview of SuperLU: algorithms, implementation, and user interface. ACM Trans. Math. Softw. 31(3), 302–325 (2005)
Li, X.S., Demmel, J.W.: SuperLU\(\_\)DIST: a scalable distributed-memory sparse direct solver for unsymmetric linear systems. ACM Trans. Math. Softw. 29(9), 110–140 (2003)
Lourakis, M.: Sparse non-linear least squares optimization for geometric vision. Eur. Conf. Comput. Vis. 2, 43–56 (2010)
Nocedal, J., Wright, S.J.: Numerical Optimization. Springer, New York (1999)
Paige, C., Saunders, M.: An algorithm for sparse linear equations and sparse least squares. ACM Trans. Math. Softw. 8(1), 43–71 (1982)
Peters, G., Wilkinson, J.H.: The least squares problem and pseudo-inverses. Comput. J. 13, 309–316 (1970)
Saad, Y.: Iterative Methods for Sparse Linear Systems, 2nd edn. SIAM, Philadelphia (2000)
Schenk, O., Gärtner, K.: PARDISO User Guide (2014). http://www.pardiso-project.org/manual/manual.pdf
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Howell, G.W., Baboulin, M. (2016). LU Preconditioning for Overdetermined Sparse Least Squares Problems. In: Wyrzykowski, R., Deelman, E., Dongarra, J., Karczewski, K., Kitowski, J., Wiatr, K. (eds) Parallel Processing and Applied Mathematics. PPAM 2015. Lecture Notes in Computer Science(), vol 9573. Springer, Cham. https://doi.org/10.1007/978-3-319-32149-3_13
Download citation
DOI: https://doi.org/10.1007/978-3-319-32149-3_13
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-32148-6
Online ISBN: 978-3-319-32149-3
eBook Packages: Computer ScienceComputer Science (R0)