Skip to main content
SpringerLink
Log in

Improving directions of negative curvature in an efficient manner

  • Published:
Annals of Operations Research Aims and scope Submit manuscript

Abstract

In order to converge to second-order KKT points, second derivative information has to be taken into account. Therefore, methods for minimization satisfying convergence to second-order KKT points must, at least implicitly, compute a direction of negative curvature of an indefinite matrix. An important issue is to determine the quality of the negative curvature direction. This problem is closely related to the symmetric eigenvalue problem. More specifically we want to develop algorithms that improve directions of negative curvature with relatively little effort, extending the proposals by Boman and Murray. This paper presents some technical improvements with respect to their work. In particular, we study how to compute “good” directions of negative curvature. In this regard, we propose a new method and we present numerical experiments that illustrate its practical efficiency compared to other proposals.

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

Access this article

Log in via an institution

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

  • Bai, Z., Demmel, J., Dongarra, J., Ruhe, A., & van der Vorst, H. (Eds.). (2000). Templates for the solution of algebraic eigenvalue problems: A practical guide. Philadelphia: SIAM.

    Google Scholar 

  • Boman, E. G. (1999). Infeasibility and negative curvature in optimization. Ph.D. Thesis, Stanford University.

  • Boman, E. G., & Murray, W. (1998). An iterative approach to computing directions of negative curvature. In: Proceedings of the fifth Copper Mountain conference on iterative methods, University of Colorado.

  • Bunch, J. R., & Parlett, B. N. (1971). Direct methods for solving symmetric indefinite systems of linear equations. SIAM Journal on Numerical Analysis, 8, 639–655.

    Article  Google Scholar 

  • Bunch, J. R., Kaufman, L., & Parlett, B. N. (1976). Decomposition of a symmetric matrix. Numerical Mathematics, 27, 95–109.

    Article  Google Scholar 

  • Cheng, S. H., & Higham, N. (1998). A modified Cholesky algorithm based on a symmetric indefinite factorization. SIAM Journal on Matrix Analysis Applications, 19, 1097–1110.

    Article  Google Scholar 

  • Conn, A. R., Gould, N. I. M., & Toint, P. L. (2000). Trust-region methods. SIAM series on optimization. Philadelphia: SIAM.

    Google Scholar 

  • Davidson, E. R. (1975). The iterative calculation of a few of the lowest eigenvalues and corresponding eigenvectors of large real symmetric matrices. Journal Computational Physics, 17, 87–94.

    Article  Google Scholar 

  • Doyle, M. (2003). A barrier algorithm for large nonlinear optimization. Ph.D. Thesis, Stanford University.

  • Fiacco, A. V., & McCormick, G. P. (1990). Nonlinear programming: sequential unconstrained minimization techniques. Philadelphia: Society for Industrial and Applied Mathematics (originally published by Wiley and Sons, New York, 1968).

    Google Scholar 

  • Fletcher, R., & Freeman, T. L. (1977). A modified Newton method for minimization. Journal of Optimization Theory and Applications, 23, 357–372.

    Article  Google Scholar 

  • Forsgren, A., Gill, P. E., & Murray, W. (1995). Computing modified Newton directions using a partial Cholesky factorization. SIAM Journal on Scientific Computing, 16, 139–150.

    Article  Google Scholar 

  • Gill, P. E., & Murray, W. (1974). Newton type methods for unconstrained and linearly constrained optimization. Mathematical Programming, 7, 311–350.

    Article  Google Scholar 

  • Gill, P. E., Murray, W., & Wright, M. H. (1981). Practical optimization. New York: Academic Press.

    Google Scholar 

  • Goldfarb, D. (1980). Curvilinear path steplength algorithms for minimization which use directions of negative curvature. Mathematical Programming, 18, 31–40.

    Article  Google Scholar 

  • Golub, G., & van Loan, C. (1996). Matrix computations (3rd ed.). London: The John Hopkins University Press.

    Google Scholar 

  • Gould, N. I. M., Lucidi, S., Roma, M., & Toint, P. L. (2000). Exploiting negative curvature directions in linesearch methods for unconstrained optimization. Optimization Methods and Software, 14, 75–98.

    Article  Google Scholar 

  • Hager, W. W., & Zhang, H. (2005). A new conjugate gradient method with guaranteed descent and efficient line search. SIAM Journal on Optimization, 16, 170–192.

    Article  Google Scholar 

  • Jacobi, C. G. J. (1846). Ueber ein leichtes Verfahrem, die in der Theorie der Säcularstörungen vorkommenden Gleichungen numerisch aufzulösen. Journal on Reine Angewandte Mathematik, 30, 51–94.

    Article  Google Scholar 

  • Lucidi, S., Rochetich, F., & Roma, M. (1998). Curvilinear stabilization techniques for truncated Newton methods in large scale unconstrained optimization. SIAM Journal on Optimization, 8, 916–939.

    Article  Google Scholar 

  • McCormick, G. (1977). A modification of Armijo’s step-size rule for negative curvature. Mathematical Programming, 13, 111–115.

    Article  Google Scholar 

  • Moguerza, J. M., & Prieto, F. J. (2003a). An Augmented-Lagrangian interior-point method using directions of negative curvature. Mathematical Programming, 95, 573–616.

    Article  Google Scholar 

  • Moguerza, J. M., & Prieto, F. J. (2003b). Combining search directions using gradient flows. Mathematical Programming, 96, 529–559.

    Article  Google Scholar 

  • Moré, J. J., & Sorensen, D. C. (1979). On the use of directions of negative curvature in a modified Newton method. Mathematical Programming, 16, 1–20.

    Article  Google Scholar 

  • Mukai, H., & Polak, E. (1978). A second order algorithm for the general nonlinear programming problem. Journal of Optimization Theory and Applications, 4, 26, 515–532.

    Article  Google Scholar 

  • Neumaier, A. (1997). On satisfying second-order optimality conditions using modified Cholesky factorizations (Tech. report). Universitat Wien, Vienna, Austria.

  • Olivares, A., Moguerza, J. M., & Prieto, F. J. (2008, accepted). Nonconvex optimization using negative curvature within a modified linesearch. European Journal of Operational Research.

  • Paige, C. C. (1971). The computation of eigenvalues and eigenvectors of very large sparse matrices. Ph.D. Thesis, London University, London, UK.

  • Pardalos, P. M. (1991). Construction of test problems in quadratic bivalent programming. ACM Transactions Mathematical Software, 17, 74–87.

    Article  Google Scholar 

  • Parlett, B. N. (1980). The symmetric eigenvalue problem. Englewood Cliffs: Prentice-Hall.

    Google Scholar 

  • Ruhe, A. (1974). Iterative eigenvalue algorithms for large symmetric matrices. International Series Numerical Mathematics, 24, 97–115.

    Google Scholar 

  • Saad, Y. (1980). On the rates of convergence of the Lanczos and the block Lanczos methods. SIAM Journal on Numerical Analysis, 17, 687–706.

    Article  Google Scholar 

  • Saad, Y. (1992). Numerical methods for large eigenvalue problems. Series in algorithms and arch. for adv. sci. comput. Manchester: Manchester University Press.

    Google Scholar 

  • Saad, Y., & Schultz, M. H. (1985). Conjugate gradient-like algorithms for solving nonsymmetric linear systems. Mathematical Computation, 44, 417–424.

    Article  Google Scholar 

  • Shultz, G. A., Schnabel, R. B., & Byrd, R. H. (1985). A family of trust-region-based algorithms for unconstrained minimization with strong global convergence properties. SIAM Journal on Numerical Analysis, 22(1), 47–67.

    Article  Google Scholar 

  • Sleijpen, G. L. G., & Van der Vorst, H. A. (1995). The Jacobi-Davidson method for eigenvalue problems as an accelerated inexact Newton scheme (Technical report). Mathematical Institute, Utrecht University.

  • Sanmatías, S., & Vercher, E. (1998). A generalized conjugate gradient algorithm. Journal of Optimization Theory and Applications, 98, 489–502.

    Article  Google Scholar 

  • Sorensen, D. C., & Yang, C. (1995). Accelerating the Lanczos iteration via polynomial spectral transformations (Technical report TR97-29). Department of Computational and Applied Mathematics, Rice University, Houston, TX 77005.

  • Sun, J., Yang, X., & Chen, X. (2005). Quadratic cost flow and the conjugate gradient method. European Journal of Operational Research, 164, 104–114.

    Article  Google Scholar 

  • Varga, R. S. (1962). Matrix iterative analysis. Englewood Cliffs: Prentice-Hall.

    Google Scholar 

  • Wilkinson, J. H. (1965). The algebraic eigenvalue problem. Oxford: Clarendon Press.

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Statistics and Operational Research, University Rey Juan Carlos, C/Tulipán s/n, 28933, Móstoles (Madrid), Spain

    Alberto Olivares & Javier M. Moguerza

Authors
  1. Alberto Olivares
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Javier M. Moguerza
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Alberto Olivares.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Olivares, A., Moguerza, J.M. Improving directions of negative curvature in an efficient manner. Ann Oper Res 166, 183–201 (2009). https://doi.org/10.1007/s10479-008-0425-z

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10479-008-0425-z

Keywords

Search

Navigation