Magnus R. Hestenes

Full Article (pdf)

Magnus Hestenes discusses his career and the story of the development of the conjugate gradient method. He discusses some of his wartime work on aerial gunnery and variational problems, including time optimal control problems. He recalls how his work, at the Institute for Numerical Analysis at UCLA led to investigation of methods for the numerical solution to problems in linear algebra. This eventually led him to the first formulation of the conjugate gradient method.

Key words: variational problems, time optimal control, aerial gunnery, aerial combat, conjugate gradient algorithm, Davidon-Fletcher-Powell method

Funding Agency:  National Bureau of Standards

Time frame: 1940's, 1950's, 1960's

People: John Todd, L. W. Cohen, E. Paxson, J. Barkley Rosser, George Forsythe, Cornelius Lanczos, Gertrude Blanch, William Karush, Marvin Stein, Eduard Stiefel

Location: Applied Mathematics Group, Columbia University; Institute for Numerical Analysis, UCLA; Rand Corporation, REAC (electronic analog computer)

Copyright: Reprinted from "A History of Scientific Computing," Stephen G.
Nash, editor, pp. 167 -179. (c) 1990 ACM Press (Association for Computing
Machinery, Inc.) by permission.