RT Journal Article
T1 An augmented Lagrangian interior-point method using directions of negative curvature
A1 Moguerza, Javier M.
A1 Prieto, Francisco J.
AB We describe an efficient implementation of an interior-point algorithm for non-convex problemsthat uses directions of negative curvature. These directions should ensure convergence to second-order KKTpoints and improve the computational efficiency of the procedure. Some relevant aspects of the implementationare the strategy to combine a direction of negative curvature and a modified Newton direction, andthe conditions to ensure feasibility of the iterates with respect to the simple bounds. The use of multivariatebarrier and penalty parameters is also discussed, as well as the update rules for these parameters.We analyzethe convergence of the procedure; both the linesearch and the update rule for the barrier parameter behaveappropriately. As the main goal of the paper is the practical usage of negative curvature, a set of numericalresults on small test problems is presented. Based on these results, the relevance of using directions of negativecurvature is discussed.
PB Springer
SN 0025-5610‎
YR 2003
FD 2003-03
LK https://hdl.handle.net/10016/14940
UL https://hdl.handle.net/10016/14940
LA eng
NO The original publication is available at  www.springerlink.com
NO Research supported by Spanish MEC grant TIC2000-1750-C06-04; Research supported by Spanish MEC grant BEC2000-0167
DS e-Archivo
RD 5 may. 2024