S. Habibi (Birmingham Uni) An interior-point method for Lasserre relaxations of binary quadratic ...

The aim of this presentation is to solve linear semidefinite programs arising from Lasserre relaxations of binary quadratic optimization problems. For this, we use an interior point method with a preconditioned conjugate gradient method to solve the linear systems. The preconditioner utilizes the low-rank structure of the solution of the relaxations. In order to fully utilize this, we re-write the moment relaxations. To treat the arising linear equality constraints, we use an $\ell_1$-penalty approach within the interior-point solver. The efficiency is demonstrated by numerical experiments and comparison with a state-of-the-art semidefinite solver.

03.04.2023, 16.00 - 17.00, Room B-365, Faculty of Civil Engineering, Thákurova 7, 166 29 Prague 6