Primal-Dual Interior-Point Methods for Domain-Driven Formulations
Abstract
We study infeasible-start primal-dual interior-point methods for convex optimization problems given in a typically natural form we denote as Domain-Driven formulation. Our algorithms extend many advantages of primal-dual interior-point techniques available for conic formulations, such as the current best complexity bounds, and more robust certificates of approximate optimality, unboundedness, and infeasibility, to Domain-Driven formulations. The complexity results are new for the infeasible-start setup used, even in the case of linear programming. In addition to complexity results, our algorithms aim for expanding the applications of, and software for interior-point methods to wider classes of problems beyond optimization over symmetric cones.
Cite
@article{arxiv.1804.06925,
title = {Primal-Dual Interior-Point Methods for Domain-Driven Formulations},
author = {Mehdi Karimi and Levent Tunçel},
journal= {arXiv preprint arXiv:1804.06925},
year = {2019}
}
Comments
44 pages, 2 figures, to appear in Mathematics of Operations Research