Applications of the alternating direction method of multipliers to the semidefinite inverse quadratic eigenvalue problem with a partial eigenstructure

Abstract

This paper shows that the alternating direction method of multipliers (ADMM) is efficient for solving the semidefinite inverse quadratic eigenvalue problem (SDIQEP) with a partial eigenstructure. We derive several ADMM-based iterative schemes for SDIQEP and demonstrate their efficiency for large-scale cases of SDIQEP numerically. © 2013 IOP Publishing Ltd.