Global convergence result for conjugate gradient methods. (English) Zbl 0794.90063

Hu, Y. F.; Storey, C.

J. Optimization Theory Appl. 71, No. 2, 399-405 (1991).

Summary: Conjugate gradient optimization algorithms depend on the search directions, \(s^{(1)}= -g^{(1)}\), \(s^{(k+1)}=- g^{(k+1)}+ \beta^{(k)}s^{(k)}\), \(k\geq 1\), with different methods arising from different choices for the scalar \(\beta^{(k)}\). In this note, conditions are given on \(\beta^{(k)}\) to ensure global convergence of the resulting algorithms.

MSC:

90C30

Nonlinear programming

Keywords:

conjugate gradient optimization; global convergence

References:

[1]	Al-Baali, M.,Descent Property and Global Convergence of the Fletcher-Reeves Method with Inexact Line Searches, IMA Journal of Numerical Analysis, Vol. 5, No. 1, pp. 121-124, 1985. · Zbl 0578.65063
[2]	Powell, M. J. D.,Nonconvex Minimization Calculations and the Conjugate Gradient Method, Report No. DAMTP 1983/NA14, Department of Applied Mathematics and Theoretical Physics, University of Cambridge, Cambridge, England, 1983. · Zbl 0531.65035
[3]	Touati-Ahmed, D., andStorey, C.,Globally Convergent Hybrid Conjugate Gradient Methods, Journal of Optimization Theory and Applications, Vol. 64, No. 2, pp. 379-397, 1990. · Zbl 0687.90081
[4]	Gilbert, J. C., andNocedal, J.,Global Convergence Properties of Conjugate Gradient Methods for Optimization, Rapport de Recherche No. 1268, Institut National de Recherche en Informatique et Automatique, Domaine de Voluceau, Rocquencourt, Le Chesnay, France, 1990.