%A Steven P. Dirkse
%A Michael C. Ferris
%T Preprocessing Techniques for Large-Scale Complementarity Problems
%D December 1995
%R 95-22
%I COMPUTER SCIENCES DEPARTMENT, UNIVERSITY OF WISCONSIN
%C MADISON, WI
%X Most Newton-based solvers for complementarity problems converge rapidly
to a solution once they are close to the solution point and the
correct active set has been found.
We discuss the design and implementation of projected
gradient and projected Newton preprocessing techniques designed to
compute a good active set quickly. Computational results
obtained using these preprocessors with PATH and SMOOTH, state-of-the-art
complementarity solvers, are given, demonstrating the value of the
projected Newton technique in this context.