Numerical experience with the truncated Newton method for unconstrained optimization (Q1090623)

The truncated Newton algorithm was devised by \textit{R. S. Dembo} and \textit{T. Steihaug} [Math. Program. 26, 190-212 (1983; Zbl 0523.90078)] for solving large sparse unconstrained optimization problems. When far from a minimum, an accurate solution to the Newton equations may not be justified. Dembo's method solves these equations by the conjugate direction method, but truncates the iteration when a required degree of accuracy has been obtained. We present favorable numerical results obtained with the algorithm and compare them with existing codes for large scale optimization.