Abstract
When a classical barrier method is applied to the solution of a nonlinear programming problem with inequality constraints, the Hessian matrix of the barrier function becomes increasingly ill-conditioned as the solution is approached. As a result, it may be desirable to consider alternative numerical algorithms. We compare the performance of two methods motivated by barrier functions. The first is a stabilized form of the classical barrier method, where a numerically stable approximation to the Newton direction is used when the barrier parameter is small. The second is a modified barrier method where a barrier function is applied to a shifted form of the problem, and the resulting barrier terms are scaled by estimates of the optimal Lagrange multipliers. The condition number of the Hessian matrix of the resulting modified barrier function remains bounded as the solution to the constrained optimization problem is approached. Both of these techniques can be used in the context of a truncated- Newton method, and hence can be applied to large problems, as well as on parallel computers. In this paper, both techniques are applied to problems with bound constraints and we compare their practical behavior.
The authors were partially supported by National Science Foundation grant DDM-9104670. R. Polyak was partially supported by NASA contract NAG3-1397 and National Science Foundation grant DMS-9300962.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
B. M. Averick, R. G. Carter, and J. J. Moré, (1991), “The Minpack-2 Test Problem Collection (Preliminary Version),” Report ANL/MCS-TM-150, Mathematics and Computer Science Division, Argonne National Laboratory, Argonne IL.
A. Ben-Tal, I. Yuzefovich and M. Zibulevsky, (1992), “Penalty/Barrier Multiplier Methods for minimax and constrained smooth convex problems,” Research Report 9/92, Optimization Laboratory, Faculty of Industrial Engineering and Management, Technion (Israel Institute of Technology).
M. G. Breitfeld and D. F. Shanno, (1993), “Preliminary computational experience with modified log-barrier functions for large-scale nonlinear programming,” in Large Scale Optimization: State of the Art, ( W. W. Hager, D. W. Hearn and P.M. Pardalos, Eds.), Kluwer Academic Publishers B. V..
A. R. Conn, N. I. M. Gould, and P. L. Toint, (1988), “Testing a class of methods for solving minimization problems with simple bounds on the variables,” Math. Comp., 50, 399–430.
A. V. Fiacco and G. P. McCormick, (1968), Nonlinear Programming: Sequential Unconstrained Minimization Techniques, John Wiley and Sons, New York ( This book has been reprinted by SIAM, Philadelphia, 1990 ).
P. E. Gill, W. Murray, and M. H. Wright, (1981), Practical Optimization, Academic Press, New York.
D. L. Jensen and R. Polyak, (1992), “The convergence of a modified barrier method for convex programming,” Technical Report 18570, IBM Thomas J. Watson Research Center, Yorktown Heights, NY.
W. Murray, (1971), “Analytical expressions for the eigenvalues and eigenvectors of the Hessian matrices of barrier and penalty functions,” J. Opt. Th. App., 7, 189–196.
W. Murray and M. H. Wright, (1976), “Efficient linear search algorithms for the logarithmic barrier function,” Report SOL 76–18, Dept. of Operations Research, Stanford University, Stanford CA.
S. G. Nash, (1993), “A parallel barrier algorithm for large-scale bound-constrained optimization,” Technical Report 93–01, Operations Research and Engineering Department, George Mason University, Fairfax VA.
S. G. Nash and J. Nocedal, (1991), “A Numerical Study of the Limited Memory BFGS Method and the Truncated-Newton Method for Large Scale Optimization,” SIAM J. Opt, 1, 358–372.
S. G. Nash and A. Sofer, (1990), “Assessing a search direction within a truncated- Newton method,” Operations Research Letters, 9, 219–221.
S. G. Nash and A. Sofer, (1991), “A general-purpose parallel algorithm for un-constrained optimization,” SIAM J. Opt, 1, 530–547.
S. G. Nash and A. Sofer, (1993), “A barrier method for large-scale constrained optimization,” ORSA Journal on Computing, 5, 40–53.
S. G. Nash and A. Sofer, (1993), “Preconditioning of reduced matrices,” Technical Report, Operations Research and Engineering Department, George Mason University, Fairfax VA.
J. M. Ortega and W. C. Rheinboldt, (1970), Iterative Solution of Nonlinear Equations in Several Variables, Academic Press, London and New York.
R. Polyak, (1992), “Modified barrier functions (theory and methods),” Mathematical Programming, 177–222.
M. H. Wright, (1992), “Determining subspace information from the Hessian of the barrier function,” Numerical Analysis Manuscript 92–02, AT& T Bell Laboratories, Murray Hill NJ.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1994 Kluwer Academic Publishers
About this chapter
Cite this chapter
Nash, S.G., Polyak, R., Sofer, A. (1994). A Numerical Comparison of Barrier and Modified Barrier Methods For Large-Scale Bound-Constrained Optimization. In: Hager, W.W., Hearn, D.W., Pardalos, P.M. (eds) Large Scale Optimization. Springer, Boston, MA. https://doi.org/10.1007/978-1-4613-3632-7_16
Download citation
DOI: https://doi.org/10.1007/978-1-4613-3632-7_16
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4613-3634-1
Online ISBN: 978-1-4613-3632-7
eBook Packages: Springer Book Archive