Showing 1–1 of 1 results for author: Kavand, A

Barrier and penalty methods for low-rank semidefinite programming with application to truss topology design

Authors: Soodeh Habibi, Arefeh Kavand, Michal Kocvara, Michael Stingl

Abstract: The aim of this paper is to solve large-and-sparse linear Semidefinite Programs (SDPs) with low-rank solutions. We propose to use a preconditioned conjugate gradient method within second-order SDP algorithms and introduce a new efficient preconditioner fully utilizing the low-rank information. We demonstrate that the preconditioner is universal, in the sense that it can be efficiently used within… ▽ More The aim of this paper is to solve large-and-sparse linear Semidefinite Programs (SDPs) with low-rank solutions. We propose to use a preconditioned conjugate gradient method within second-order SDP algorithms and introduce a new efficient preconditioner fully utilizing the low-rank information. We demonstrate that the preconditioner is universal, in the sense that it can be efficiently used within a standard interior-point algorithm, as well as a newly developed primal-dual penalty method. The efficiency is demonstrated by numerical experiments using the truss topology optimization problems of growing dimension. △ Less

Submitted 18 May, 2021; originally announced May 2021.

MSC Class: 90C22; 90C51; 65F08; 74P05