Accession Number : AD0746479

Title :   Global Convergence for Newton Methods in Mathematical Programming,

Corporate Author : TEXAS UNIV AUSTIN CENTER FOR NUMERICAL ANALYSIS

Personal Author(s) : Daniel,James W.

Report Date : JUN 1972

Pagination or Media Count : 14

Abstract : In constrained optimization problems in mathematical programming one wants to minimize a functional f(x) over a given set C. If, at an approximate solution (x sub n), one replaces f(x) by its Taylor series expansion through quadratic terms at (x sub n) and denotes by x sub (n+1) the minimizing point for this over C, one has a direct analogue of Newton's method. The local convergence of this has been previously analyzed; here the author gives global convergence results for this and the similar algorithm in which the constraint set C is also linearized at each step. (Author)

Descriptors :   (*MATHEMATICAL PROGRAMMING, ITERATIONS), CONVERGENCE, CONVEX SETS, APPROXIMATION(MATHEMATICS), ALGORITHMS

Subject Categories : Operations Research

Distribution Statement : APPROVED FOR PUBLIC RELEASE