Using a Factored Dual in Augmented Lagrangian Methods for Semidefinite Programming
Preprint
Abstract
In the context of augmented Lagrangian approaches for solving semidefinite programming problems, we investigate the possibility of eliminating the positive semidefinite constraint on the dual matrix by employing a factorization. Hints on how to deal with the resulting unconstrained maximization of the augmented Lagrangian are given. We further use the approximate maximum of the augmented Lagrangian with the aim of improving the convergence rate of alternating direction augmented Lagrangian frameworks. Numerical results are reported, showing the benefits of the approach.Keywords
All Related Versions
This publication has 9 references indexed in Scilit:
- The direct extension of ADMM for multi-block convex minimization problems is not necessarily convergentMathematical Programming, 2014
- Semidefinite relaxations of ordering problemsMathematical Programming, 2013
- Alternating direction augmented Lagrangian methods for semidefinite programmingMathematical Programming Computation, 2010
- Regularization Methods for Semidefinite ProgrammingSIAM Journal on Optimization, 2009
- A Boundary Point Method to Solve Semidefinite ProgramsComputing, 2006
- Local Minima and Convergence in Low-Rank Semidefinite ProgrammingMathematical Programming, 2004
- Solving Large Scale Semidefinite Programs via an Iterative Solver on the Augmented SystemsSIAM Journal on Optimization, 2004
- A nonlinear programming algorithm for solving semidefinite programs via low-rank factorizationMathematical Programming, 2003
- Constrained Optimization and Lagrange Multiplier Methods1982