×

Mise a jour de la métrique dans les méthodes de quasi-Newton réduites en optimisation avec contraintes d’égalité. (Updating for the metric in the reduced quasi-Newton methods in optimization with equality constraints). (French) Zbl 0657.65087

This paper is concerned with local superlinear convergence of reduced Hessian approximation methods for equality-constrained optimization. Conditions for obtaining superlinear convergence, instead of 2-step superlinear convergence, are discussed. Variations on different reduced Hessian methods are proposed and analyzed. Numerical comparisons are not provided.
Reviewer: T.F.Coleman

MSC:

65K05 Numerical mathematical programming methods
65H10 Numerical computation of solutions to systems of equations
90C30 Nonlinear programming
PDFBibTeX XMLCite
Full Text: DOI EuDML

References:

[1] L. ARMIJO (1966. Minimization of functions having Lipschitz continuous first partial derivatives. Pacific Journal of Mathematics 16/1,1-3. Zbl0202.46105 MR191071 · Zbl 0202.46105 · doi:10.2140/pjm.1966.16.1
[2] J BLUM, J. Ch. GILBERT, B. THOORIS (1985). Parametric identification of the plasma current density from the magnetic measurements and the pressure profile, code IDENTC Report of JET contract number JT3/9008.
[3] J. F. BONNANS, D. GABAY (1984. Une éxtension de la programmation quadratique successive. Lecture Notes in Control and Information Sciences 63, 16-31. A. Bensoussan, J. L Lions (eds). Springer-Verlag. Zbl0559.90081 MR876712 · Zbl 0559.90081
[4] C. G. BROYDEN (1969). A new double-rank minimization algorithm. Notices of the American Mathematical Society 16, 670. · Zbl 0164.45101
[5] C G. BROYDEN, J. E. DENNIS, J. J. MORE (1973). On the local and superlinear convergence of quasi-Newton methods. Journal of the Institute of Mathematics and its Applications 12, 223-245 Zbl0282.65041 MR341853 · Zbl 0282.65041 · doi:10.1093/imamat/12.3.223
[6] R. H. BYRD (1985). An example of irregular convergence in some constrained optimization methods that use the projected hessian. Mathematical Programming 32, 232-237. Zbl0576.90079 MR793692 · Zbl 0576.90079 · doi:10.1007/BF01586093
[7] R. H. BYRD, R. B. SCHNABEL (1986). Continuity of the null space basis and constrained optimization. Mathematical Programming 35, 32-41. Zbl0598.90072 MR842632 · Zbl 0598.90072 · doi:10.1007/BF01589439
[8] T F COLEMAN, A. R. CONN (1982 a). Nonlinear programming via an exact penalty function: asymptotic analysis. Mathematical Programming 24, 123-136. Zbl0501.90078 MR674627 · Zbl 0501.90078 · doi:10.1007/BF01585100
[9] T. F. COLEMAN, A. R. CONN (1982 b). Nonlinear programming via an exact penalty function: global analysis. Mathematical Programming 24, 137-161. Zbl0501.90077 MR674628 · Zbl 0501.90077 · doi:10.1007/BF01585101
[10] T. F. COLEMAN, A. R. CONN (1984. On the local convergence of a quasi-Newton method for the nonlinear programming problem. SIAM Journal on Numerical Analysis 21/4, 755-769. Zbl0566.65046 MR749369 · Zbl 0566.65046 · doi:10.1137/0721051
[11] J. E. DENNIS, J. J. MORE (1974) A characterization of superlinear convergence and its application to quasi-Newton methods Mathematics of Computation 28/126, 549-560. Zbl0282.65042 MR343581 · Zbl 0282.65042 · doi:10.2307/2005926
[12] J. E. DENNIS, J. J. MORE (1977). Quasi-Newton methods, motivation and theory. SIAM Review 19, 46-89. Zbl0356.65041 MR445812 · Zbl 0356.65041 · doi:10.1137/1019005
[13] R. FLETCHER (1970). A new approach to variable metric algorithms. Journal 13/3, 317-322. Zbl0207.17402 · Zbl 0207.17402 · doi:10.1093/comjnl/13.3.317
[14] R. FLETCHER (1981). Practical Methods of Optimization Vol. 2 : Constrained Optimization. John Wiley & Sons. Zbl0474.65043 MR633058 · Zbl 0474.65043
[15] D. GABAY (1982a). Minimizing a differentiable function over a differential manifold. Journal of Optimization Theory and Applications 37/2, 177-219. Zbl0458.90060 MR663521 · Zbl 0458.90060 · doi:10.1007/BF00934767
[16] D. GABAY (1982b). Reduced quasi-Newton methods with feasibility improvement for nonlinearly constrained optimization. Mathematical Programming Study 16,18-44. Zbl0477.90065 MR650627 · Zbl 0477.90065 · doi:10.1007/BFb0120946
[17] R. P. GE, M. J. D. POWELL (1983). The convergence of variable metric matrices in unconstrained optimization Mathematical Programming 27, 123-143. Zbl0532.49015 MR718055 · Zbl 0532.49015 · doi:10.1007/BF02591941
[18] J. Ch. GILBERT (1986a). Une méthode à métrique variable réduite en optimisation avec contraintes d’égalité non linéaires Rapport de recherche de l’INRIA RR-482, 78153 Le Chesnay Cedex, France.
[19] J. Ch. GILBERT (1986b). On the local and global convergence of a reduced quasi-Newton method Rapport de recherche de l’INRIA RR-565, 78153 Le Chesnay Cedex, France (version révisée dans IIASA Workmg Paper WP-87-113). Zbl0676.90061 · Zbl 0676.90061 · doi:10.1080/02331938908843462
[20] J. Ch. GILBERT (1986b). Une méthode de quasi-Newton réduite en optimisation sous contraintes avec priorité à la restauration. Lecture Notes in Control and Information Sciences 83, 40-53. A. Bensoussan, J. L. Lions (eds), Sprmger-Verlag. Zbl0599.90112 MR870388 · Zbl 0599.90112
[21] J. Ch. GILBERT (-) (en préparation).
[22] D. GOLDFARB (1970). A family of variable metric methods derived by variational means. Mathematics of Computation 24, 23-26. Zbl0196.18002 MR258249 · Zbl 0196.18002 · doi:10.2307/2004873
[23] S. P. HAN (1976). Superlinearly convergent variable metric algorithms for general nonlinear programming problems. Mathematical Programming 11, 263-282. Zbl0364.90097 MR483440 · Zbl 0364.90097 · doi:10.1007/BF01580395
[24] S. P. HAN (1977). A globally convergent method for nonlinear programming. Journal of Optimization Theory and Applications 22/3, 297-309. Zbl0336.90046 MR456497 · Zbl 0336.90046 · doi:10.1007/BF00932858
[25] D. Q. MAYNE, E. POLAK (1982). A superlinearly convergent algorithm for constrained optimization problems. Mathematical Programming Study 16, 45-61. Zbl0477.90071 MR650628 · Zbl 0477.90071 · doi:10.1007/BFb0120947
[26] H. MUKAI, E. POLAK (1978). On the use of approximations in algorithms for optimization problems with equality and inequality constraints. SIAM Journal on Numerical Analysis 15/4, 674-693. Zbl0392.49017 MR497967 · Zbl 0392.49017 · doi:10.1137/0715045
[27] J. NOCEDAL, M. L. OVERTON (1985). Projected Hessian updating algorithms for nonlinearly constrained optimization. SIAM Journal on Numerical Analysis 22/5, 821-850. Zbl0593.65043 MR799115 · Zbl 0593.65043 · doi:10.1137/0722050
[28] M. J. D. POWELL (1971). On the convergence of the variable metric algorithm. Journal of the Institute of Mathematics and its Applications 7, 21-36. Zbl0217.52804 MR279977 · Zbl 0217.52804 · doi:10.1093/imamat/7.1.21
[29] M. J. D. POWELL (1976). Some global convergence properties of a variable metric algorithm for minimization without exact line searches. Nonlinear Programming, SIAM-AMS Proceedings, Vol. 9, American Mathematical Society, Providence, R.I. Zbl0338.65038 MR426428 · Zbl 0338.65038
[30] M. J. D. POWELL(1978). The convergence of variable metric methods for nonlinearly constrained optimization calculations. Nonlinear Programming 3, 27-63. O. L. Mangasarian, R. R. Meyer, S. M. Robinson (eds), Academic Press, New York. Zbl0464.65042 MR507858 · Zbl 0464.65042
[31] D. F. SHANNO (1970). Conditioning of quasi-Newton methods for function minimization. Mathematics of Computation 24, 647-656. Zbl0225.65073 MR274029 · Zbl 0225.65073 · doi:10.2307/2004840
[32] R. B. WILSON (1963). A simplicial algorithm for concave programming. Ph. D. Thesis. Graduate School of Business Administration, Havard Univ., Cambridge, MA.
[33] Y. YUAN (1985). An only 2-step Q-superlinear convergence example for some algonthms that use reduced Hessian approximations Mathematical Programming 32, 224-231. Zbl0565.90060 MR793691 · Zbl 0565.90060 · doi:10.1007/BF01586092
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.