Friday, 11:30 - 11:55 h, Room: H 1012


Jean-Louis Goffin
Solving unconstrained nonconvex programs with ACCPM

Coauthors: Ahad Dehghani, Dominique Orban


We suggest the use of ACCPM and proximal ACCPM, well known techniques for convex
programming problems, in a sequential convex programming method
based on ACCPM and convexification techniques to tackle unconstrained
problems with a non-convex objective function, by adding a proximal term to the objectiver. We also report a comparison of our
method with some existing algorithms: the steepest descent method and nonlinear conjugate gradient algorithms.
We use a sequence of convex functions and show that the global minimizers of these convex functions converge to a local minimizer of the original nonconvex objective function f. These convex functions are minimized by using ACCPM-prox, a code developed by J. P. Vial.
We use the set problem CUTEr and tested two version of our algorithm, ACCPM\AdaotTol and ACCPM\FixTol on 158 problems of this set. The number of variables on these 158 problems varies from 2 to 20,000. This software presents ACCPM\AdaptTol as the best solver on more that 22% of the problems and it can solve approximately 85% of the problems.


Talk 3 of the contributed session Fri.1.H 1012
"Topics in nonsmooth nonconvex optimization" [...]
Cluster 17
"Nonsmooth optimization" [...]


  pay day loans . What can cause long-term use of Viagra Sale? In the network and other sources of information, there is no reliable data on the long-term use of Viagra and its negative effects on the body.