Title: Modified polynomial-time full-NT-step infeasible interior-point algorithm for symmetric optimisation
Authors: Maryam Zangiabadi; Soodabeh Asadi; Hossein Mansouri
Addresses: Department of Applied Mathematics, Faculty of Mathematical Sciences, Shahrekord University, P.O. Box 115, Shahrekord, Iran ' Department of Applied Mathematics, Faculty of Mathematical Sciences, Shahrekord University, P.O. Box 115, Shahrekord, Iran ' Department of Applied Mathematics, Faculty of Mathematical Sciences, Shahrekord University, P.O. Box 115, Shahrekord, Iran
Abstract: In this paper, we improve the full Nesterov-Todd (NT)-step infeasible-interior-point algorithm for symmetric optimisation of Gu et al. (2011). Our proposed algorithm differs from Gu et al.'s algorithm such that it has the advantage that no centring steps are needed. In each main iteration of Gu et al.'s algorithm, one feasibility and a few centring steps are needed to get feasible iterates for a pair of perturbed problems, close enough to its central path. In this paper, we perform only one full-NT-step in each main iteration of the algorithm. Despite eliminating the centring steps, the complexity bound of our algorithm matches the best obtained one for symmetric optimisation.
Keywords: symmetric optimisation; Euclidean Jordan algebras; polynomial complexity; infeasible interior-point methods; Nesterov-Todd.
International Journal of Operational Research, 2017 Vol.28 No.2, pp.290 - 306
Received: 23 Sep 2014
Accepted: 17 Dec 2014
Published online: 10 Jan 2017 *