A family of variable-metric methods derived by variational means

Goldfarb, Donald

doi:10.1090/S0025-5718-1970-0258249-6

A family of variable-metric methods derived by variational means
HTML articles powered by AMS MathViewer

by Donald Goldfarb PDF

Math. Comp. 24 (1970), 23-26 Request permission

Abstract:

A new rank-two variable-metric method is derived using Greenstadt’s variational approach [Math. Comp., this issue]. Like the Davidon-Fletcher-Powell (DFP) variable-metric method, the new method preserves the positive-definiteness of the approximating matrix. Together with Greenstadt’s method, the new method gives rise to a one-parameter family of variable-metric methods that includes the DFP and rank-one methods as special cases. It is equivalent to Broyden’s one-parameter family [Math. Comp., v. 21, 1967, pp. 368–381]. Choices for the inverse of the weighting matrix in the variational approach are given that lead to the derivation of the DFP and rank-one methods directly.

References

C. G. Broyden, Quasi-Newton methods and their application to function minimisation, Math. Comp. 21 (1967), 368–381. MR 224273, DOI 10.1090/S0025-5718-1967-0224273-2

Variable Metric Method for Minimization

William C. Davidon, Variance algorithm for minimization, Comput. J. 10 (1967/68), 406–410. MR 221738, DOI 10.1093/comjnl/10.4.406
R. Fletcher and M. J. D. Powell, A rapidly convergent descent method for minimization, Comput. J. 6 (1963/64), 163–168. MR 152116, DOI 10.1093/comjnl/6.2.163
D. Goldfarb, Sufficient conditions for the convergence of a variable metric algorithm, Optimization (Sympos., Univ. Keele, Keele, 1968) Academic Press, London, 1969, pp. 273–281. MR 0287905
J. Greenstadt, Variations on variable-metric methods. (With discussion), Math. Comp. 24 (1970), 1–22. MR 258248, DOI 10.1090/S0025-5718-1970-0258248-4

Another Variable Metric Method