A Coordinate Ascent Method for Solving Semidefinite Relaxations of Non-convex Quadratic Integer Programs

  • Christoph Buchheim
  • Maribel MontenegroEmail author
  • Angelika Wiegele
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9849)


We present a coordinate ascent method for a class of semidefinite programming problems that arise in non-convex quadratic integer optimization. These semidefinite programs are characterized by a small total number of active constraints and by low-rank constraint matrices. We exploit this special structure by solving the dual problem, using a barrier method in combination with a coordinate-wise exact line search. The main ingredient of our algorithm is the computationally cheap update at each iteration and an easy computation of the exact step size. Compared to interior point methods, our approach is much faster in obtaining strong dual bounds. Moreover, no explicit separation and reoptimization is necessary even if the set of primal constraints is large, since in our dual approach this is covered by implicitly considering all primal constraints when selecting the next coordinate.


Semidefinite programming Non-convex quadratic integer optimization Coordinate descent method 


  1. 1.
    Borchers, B.: CSDP, a C library for semidefinite programming. Optim. Methods Softw. 11(1–4), 613–623 (1999)MathSciNetCrossRefzbMATHGoogle Scholar
  2. 2.
    Buchheim, C., Wiegele, A.: Semidefinite relaxations for non-convex quadratic mixed-integer programming. Math. Program. 141(1–2), 435–452 (2013)MathSciNetCrossRefzbMATHGoogle Scholar
  3. 3.
    Dong, H.: Relaxing nonconvex quadratic functions by multiple adaptive diagonal perturbations. SIAM J. Optim. (accepted for publication)Google Scholar
  4. 4.
    Goemans, M.X., Williamson, D.P.: Improved approximation algorithms for maximum cut and satisfiability problems using semidefinite programming. J. ACM 42(6), 1115–1145 (1995)MathSciNetCrossRefzbMATHGoogle Scholar
  5. 5.
    Kapoor, S., Vaidya, P.M.: Fast algorithms for convex quadratic programming and multicommodity flows. In: Proceedings of the 18th Annual ACM Symposium on Theory of Computing, pp. 147–159 (1986)Google Scholar
  6. 6.
    Kozlov, M.K., Tarasov, S.P., Hačijan, L.G.: The polynomial solvability of convex quadratic programming. USSR Comput. Math. Math. Phys. 20(5), 223–228 (1980)MathSciNetCrossRefzbMATHGoogle Scholar
  7. 7.
    Lovász, L.: On the Shannon capacity of a graph. IEEE Trans. Inf. Theory 25(1), 1–7 (1979)MathSciNetCrossRefzbMATHGoogle Scholar
  8. 8.
    Pardalos, P.M., Vavasis, S.A.: Quadratic programming with one negative eigenvalue is NP-hard. J. Global Optim. 1, 15–22 (1991)MathSciNetCrossRefzbMATHGoogle Scholar
  9. 9.
    Ye, Y., Tse, E.: An extension of Karmarkar’s projective algorithm for convex quadratic programming. Math. Program. 44, 157–179 (1989)MathSciNetCrossRefzbMATHGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  • Christoph Buchheim
    • 1
  • Maribel Montenegro
    • 1
    Email author
  • Angelika Wiegele
    • 2
  1. 1.Fakultät für MathematikTechnische Universität DortmundDortmundGermany
  2. 2.Department of MathematicsAlpen-Adria-Universität KlagenfurtKlagenfurtAustria

Personalised recommendations