Skip to main content
Log in

An augmented LSQR method

  • Original Paper
  • Published:
Numerical Algorithms Aims and scope Submit manuscript

Abstract

The LSQR iterative method for solving least-squares problems may require many iterations to determine an approximate solution with desired accuracy. This often depends on the fact that singular vector components of the solution associated with small singular values of the matrix require many iterations to be determined. Augmentation of Krylov subspaces with harmonic Ritz vectors often makes it possible to determine the singular vectors associated with small singular values with fewer iterations than without augmentation. This paper describes how Krylov subspaces generated by the LSQR iterative method can be conveniently augmented with harmonic Ritz vectors. Computed examples illustrate the competitiveness of the augmented LSQR method proposed.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Baglama, J., Reichel, L.: Augmented implicitly restarted Lanczos bidiagonalization methods. SIAM J. Sci. Comput. 27, 19–42 (2005)

    Article  MathSciNet  MATH  Google Scholar 

  2. Baglama, J., Reichel, L.: Restarted block Lanczos bidiagonalization methods. Numer. Algorithms 43, 251–272 (2006)

    Article  MathSciNet  MATH  Google Scholar 

  3. Baglama, J., Reichel, L.: An implicitly restarted block Lanczos bidiagonalization method using Leja shifts. BIT Numer. Math. (in press)

  4. Benzi, M., Tuma, M.: A robust preconditioner with low memory requirements for large sparse least squares problems. SIAM J. Sci. Comput. 25, 499–512 (2003)

    Article  MathSciNet  MATH  Google Scholar 

  5. Björck, Å.: Numerical Methods for Least Squares Problems. SIAM, Philadelphia (1996)

    Book  MATH  Google Scholar 

  6. Björck, Å., Yuan, J.Y.: Preconditioners for least squares problems by LU factorization. Electron. Trans. Numer. Anal. 8, 26–35 (1997)

    Google Scholar 

  7. Boisvert, R., Pozo, R., Remington, K., Miller, B., Lipman, R.: MatrixMarket. The matrices are available at http://math.nist.gov/MatrixMarket/ (1996)

  8. Choi, S.-C.: Iterative methods for singular linear equations and least squares. Ph.D. thesis, Institute for Computational and Mathematical Engineering, Stanford University (2006)

  9. Duff, I.S., Grimes, R.G., Lewis, J.G.: User’s guide for the Harwell–Boeing sparse matrix collection (release I), France (1992). Matrices available at http://math.nist.bov/MatrixMarket/

  10. Fong, D.C.-L., Saunders, M.A.: LSMR: an iterative algorithm for sparse least-squares problems. SIAM J. Sci. Comput. 33, 2950–2971 (2011)

    Article  MathSciNet  MATH  Google Scholar 

  11. Hayami, K., Yin, J.-F., Ito, T.: GMRES methods for least squares problems. SIAM J. Matrix Anal. Appl. 31, 2400–2430 (2010)

    Article  MathSciNet  MATH  Google Scholar 

  12. Hochstenbach, M.E.: Harmonic and refined extraction methods for the singular value problem, with applications in least squares problems. BIT Numer. Math. 44, 721–754 (2004)

    Article  MathSciNet  MATH  Google Scholar 

  13. Jia, Z.: Some properties of LSQR for large sparse linear least squares problems. J. Syst. Sci. Complex. 23, 815–821 (2010)

    Article  MathSciNet  MATH  Google Scholar 

  14. Jia, Z., Niu, D.: An implicitly restarted refined bidiagonalization Lanczos method for computing a partial singular value decomposition. SIAM J. Matrix Anal. Appl. 25, 246–265 (2003)

    Article  MathSciNet  MATH  Google Scholar 

  15. Jia, Z., Niu, D.: A refined harmonic Lanczos bidiagonalization method and an implicitly restarted algorithm for computing the smallest singular triplets of large matrices. SIAM J. Sci. Comput. 32, 714–744 (2010)

    Article  MathSciNet  MATH  Google Scholar 

  16. Karimi, S., Salkuyeh, D.K., Toutounian, F.: A preconditioner for the LSQR algorithm. J. Appl. Math. Informatics 26(1–2), 213–222 (2008)

    Google Scholar 

  17. Kokiopoulou, E., Bekas, C., Gallopoulos, E.: Computing smallest singular triplets with implicitly restarted Lanczos bidiagonalization. Appl. Numer. Math. 49, 39–61 (2004)

    Article  MathSciNet  MATH  Google Scholar 

  18. Larsen, R.M.: Lanczos bidiagonalization with partial reorthogonalization. Ph.D. thesis, Dept. Computer Science, University of Aarhus, Aarhus, Denmark (1998)

  19. Larsen, R.M.: Combining implicit restarts and partial reorthogonalization in Lanczos bidiagonalization (2001). http://soi.stanford.edu/~rmunk/PROPACK/

  20. Morgan, R.B.: Computing interior eigenvalues of large matrices. Linear Algebra Appl. 154–156, 289–309 (1991)

    Article  Google Scholar 

  21. Morgan, R.B.: A restarted GMRES method augmented with eigenvectors. SIAM J. Matrix Anal. Appl. 16, 1154–1171 (1995)

    Article  MathSciNet  MATH  Google Scholar 

  22. Morgan, R.B.: Implicitly restarted GMRES and Arnoldi methods for nonsymmetric systems of equations. SIAM J. Matrix Anal. Appl. 21, 1112–1135 (2000)

    Article  MathSciNet  MATH  Google Scholar 

  23. Morgan, R.B.: GMRES with deflated restarting. SIAM J. Sci. Comput. 24, 20–37 (2002)

    Article  MathSciNet  MATH  Google Scholar 

  24. Paige, C.C.: Bidiagonalization of matrices and solution of linear equations. SIAM J. Numer. Anal. 11, 197–209 (1974)

    Article  MathSciNet  MATH  Google Scholar 

  25. Paige, C.C., Parlett, B.N., van der Vorst, H.A.: Approximate solutions and eigenvalue bounds from Krylov subspaces. Numer. Linear Algebra Appl. 2, 115–134 (1995)

    Article  MathSciNet  MATH  Google Scholar 

  26. Paige, C.C., Saunders, M.A.: LSQR: an algorithm for sparse linear equations and sparse least squares. ACM Trans. Math. Softw. 8, 43–71 (1982)

    Article  MathSciNet  MATH  Google Scholar 

  27. Reichel, L., Ye, Q.: A generalized LSQR algorithm. Numer. Linear Algebra Appl. 15, 643–660 (2008)

    Article  MathSciNet  MATH  Google Scholar 

  28. Saad, Y.: Iterative Methods for Sparse Linear Systems, 2nd edn. SIAM, Philadelphia (2003)

    Book  MATH  Google Scholar 

  29. Simon, H.D., Zha, H.: Low rank matrix approximation using the Lanczos bidiagonalization process with applications. SIAM J. Sci. Comput. 21, 2257–2274 (2000)

    Article  MathSciNet  MATH  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to J. Baglama.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Baglama, J., Reichel, L. & Richmond, D. An augmented LSQR method. Numer Algor 64, 263–293 (2013). https://doi.org/10.1007/s11075-012-9665-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11075-012-9665-8

Keywords

Mathematics Subject Classfications (2010)

Navigation