An infeasible interior-point algorithm for monotone linear complementarity problems based on a finite hyperbolic kernel function

Le résumé fourni par la source

summary:This paper concerns an infeasible kernel-based interior-point algorithm (IPA) for monotone linear complementarity problems (LCPs). Our algorithm differs from other existing algorithms in the literature since its feasibility step is induced by a finite hyperbolic barrier term. The convergence analysis shows that the proposed algorithm is well-defined and its complexity bound coincides with the currently best-known iteration bound of infeasible interior-point methods for monotone LCPs. Moreover, the practical performance of our algorithm is validated by some extensive numerical tests. To the best of our knowledge, this is the first full-Newton step infeasible IPA based on a hyperbolic kernel function for solving monotone LCPs.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.