On diagonally preconditioning the truncated Newton method for super-scale linearly constrained nonlinear prrogramming (Q800833): Difference between revisions

First, the author presents the basic ideas and algorithms of the truncated and of the preconditioning truncated Newton methods for the problem min f(x); \(x\in {\mathbb{R}}^ n\). Further, these methods are applied to linearly constrained nonlinear programming problems. The corresponding algorithms are precisely described and some useful comments, recommendations and computational results are given, concerning the deactivating strategies, preconditioning procedures, stopping tests, storage and time-consumption.