Convergence of an Infeasible Primal-Dual Interior-Point Method for ConvexProgramming

摘要

We consider the infeasible primal-dual algorithm for smooth convex programmingrecently introduced by Vial. We show, under mild assumptions, that a 'SUMT' or 'long-step path following' version of the algorithm is globally convergent. The steplength on each iteration is based on a merit function which is a modification of the potential function proposed by Vial. (Copyright (c) 1993 by Faculty of Technical Mathematics and Informatics, Delft, The Netherlands.)