Mathematical Programming

, Volume 102, Issue 3, pp 589–608

# An improved semidefinite programming relaxation for the satisfiability problem

• Miguel F. Anjos
Article

## Abstract.

The satisfiability (SAT) problem is a central problem in mathematical logic, computing theory, and artificial intelligence. An instance of SAT is specified by a set of boolean variables and a propositional formula in conjunctive normal form. Given such an instance, the SAT problem asks whether there is a truth assignment to the variables such that the formula is satisfied. It is well known that SAT is in general NP-complete, although several important special cases can be solved in polynomial time. Semidefinite programming (SDP) refers to the class of optimization problems where a linear function of a matrix variable X is maximized (or minimized) subject to linear constraints on the elements of X and the additional constraint that X be positive semidefinite. We are interested in the application of SDP to satisfiability problems, and in particular in how SDP can be used to detect unsatisfiability. In this paper we introduce a new SDP relaxation for the satisfiability problem. This SDP relaxation arises from the recently introduced paradigm of “higher liftings” for constructing semidefinite programming relaxations of discrete optimization problems. To derive the SDP relaxation, we first formulate SAT as an optimization problem involving matrices. Relaxing this formulation yields an SDP which significantly improves on the previous relaxations in the literature. The important characteristics of the SDP relaxation are its ability to prove that a given SAT formula is unsatisfiable independently of the lengths of the clauses in the formula, its potential to yield truth assignments satisfying the SAT instance if a feasible matrix of sufficiently low rank is computed, and the fact that it is more amenable to practical computation than previous SDPs arising from higher liftings. We present theoretical and computational results that support these claims.

## Keywords

Semidefinite programming Satisfiability Discrete optimization

## References

1. 1.
Anjos, M.F.: New Convex Relaxations for the Maximum Cut and VLSI Layout Problems. PhD thesis, University of Waterloo, 2001, Published online at http://etd.uwaterloo.ca/etd/manjos2001.pdf
2. 2.
Anjos, M.F., Wolkowicz, H.: Geometry of semidefinite max-cut relaxations via matrix ranks. J. Comb. Optim. 6 (3), 237–270 (2002)
3. 3.
Anjos, M.F., Wolkowicz, H.: Semidefinite programming for discrete optimization and matrix completion problems. Discr. App. Math. 123 (1–2), 513–577 (2002)Google Scholar
4. 4.
Anjos, M.F., Wolkowicz, H.: Strengthened semidefinite relaxations via a second lifting for the max-cut problem. Discr. App. Math. 119 (1–2), 79–106 (2002)Google Scholar
5. 5.
Balas, E., Ceria, S., Cornuéjols, G.: A lift-and-project cutting plane algorithm for mixed 0–1 programs. Math. Programming 58 (3, Ser. A), 295–324 (1993)Google Scholar
6. 6.
Bauschke, H.H. Kruk, S.G.: The method of reflection-projection for convex feasibility problems with an obtuse cone. Technical report, Oakland University, Rochester MI, U.S.A., February 2002Google Scholar
7. 7.
Benson, S.J., Ye, Y., Zhang, X.: Solving large-scale sparse semidefinite programs for combinatorial optimization. SIAM J. Optim. 10 (2), 443–461 (electronic), (2000)
8. 8.
Bienstock, D., Zuckerberg, M.: Subset algebra lift operators for 0-1 integer programming. Technical Report CORC 2002-01, Columbia University, July 2002Google Scholar
9. 9.
Burer, S.: Semidefinite programming in the space of partial positive semidefinite matrices. Technical report, Department of Management Sciences, University of Iowa, May 2002Google Scholar
10. 10.
Conforti, M., Cornuéjols, G.: A class of logic problems solvable by linear programming. J. Assoc. Comput. Mach. 42 (5), 1107–1113 (1995)
11. 11.
de Klerk, E., van Maaren, H.: On semidefinite programming relaxations of (2+p)-SAT. Ann. Math. Artificial Intelligence 37 (3), 285–305 (2003)
12. 12.
de Klerk, E., van Maaren, H., Warners, J.P.: Relaxations of the satisfiability problem using semidefinite programming. J. Automat. Reason. 24 (1–2), 37–65 (2000)Google Scholar
13. 13.
Ferris, M., Mesnier, M., Moré, J.: NEOS and Condor: Solving optimization problems over the Internet. ACM Trans. Math. Softw. 26 (1), 1–18 (2000)Google Scholar
14. 14.
Fukuda, M., Kojima, M., Murota, K., Nakata, K.: Exploiting sparsity in semidefinite programming via matrix completion I: General framework. SIAM J. Optim., 11 3 (01), 647–674 (electronic), (2000)Google Scholar
15. 15.
Goemans, M.X., Williamson, D.P.: New Open image in new windowapproximation algorithms for the maximum satisfiability problem. SIAM J. Disc. Math. 7 (4), 656–666 (1994)Google Scholar
16. 16.
Goemans, M.X., Williamson, D.P.: Improved approximation algorithms for maximum cut and satisfiability problems using semidefinite programming. J. Assoc. Comput. Mach. 42 (6), 1115–1145 (1995)
17. 17.
Grigoriev, D., Hirsch, E.A., Pasechnik, D.V.: Complexity of semialgebraic proofs. Moscow Math. J. 2 (4), 647–679 (2002)
18. 18.
Gu, J.: Global optimization for satisfiability (SAT) problem. IEEE Trans. on Know. and Data Eng. 6 (3), 361–381 (1994)Google Scholar
19. 19.
Gu, J., Purdom, P.W., Franco, J., Wah, B.W.: Algorithms for the satisfiability (SAT) problem: a survey. In: Satisfiability problem: theory and applications (Piscataway, NJ, 1996), Amer. Math. Soc., Providence, RI, 1997, pp. 19–151Google Scholar
20. 20.
Halperin, E., Zwick, U.: Approximation algorithms for MAX 4-SAT and rounding procedures for semidefinite programs. J. Algorithms 40 (2), 184–211 (2001)
21. 21.
Helmberg, C.: A cutting plane algorithm for large scale semidefinite relaxations. Technical Report ZR-01-26, Konrad-Zuse-Zentrum Berlin, October 2001. To appear in the Padberg Festschrift “The Sharpest Cut”, SIAMGoogle Scholar
22. 22.
Helmberg, C., Kiwiel, K.C.: A spectral bundle method with bounds. Math. Program. 93 (2, Ser. A), 173–194 (2002)Google Scholar
23. 23.
Helmberg, C., Rendl, F.: A spectral bundle method for semidefinite programming. SIAM J. Optim. 10 (3), 673–696 (electronic), (2000)
24. 24.
Henrion, D., Lasserre, J.B.: Gloptipoly - global optimization over polynomials with matlab and sedumi. Technical report, LAAS-CNRS, Toulouse, France, 2002Google Scholar
25. 25.
Karloff, H., Zwick, U.: A 7/8-approximation algorithm for MAX 3SAT? In: Proc. of 38th FOCS, 1997, pp. 406–415Google Scholar
26. 26.
Kočvara, M., Stingl, M.: PENNON - a generalized augmented lagrangian method for semidefinite programming. Technical Report 286, Institute of Applied Mathematics, University of Erlangen, 2001Google Scholar
27. 27.
Kočvara, M., Stingl, M.: PENNON - a code for convex nonlinear and semidefinite programming. Technical Report 290, Institute of Applied Mathematics, University of Erlangen, 2002Google Scholar
28. 28.
Lasserre, J.B.: Optimality conditions and LMI relaxations for 0-1 programs. Technical report, LAAS-CNRS, Toulouse, France, 2000Google Scholar
29. 29.
Lasserre, J.B.: Global optimization with polynomials and the problem of moments. SIAM J. Optim. 11 (3), 796–817 (electronic), (2000/01)
30. 30.
Lasserre, J.B.: An explicit equivalent positive semidefinite program for nonlinear 0-1 programs. SIAM J. Optim. 12 (3), 756–769 (electronic), (2002)
31. 31.
Laurent, M.: Semidefinite relaxations for max-cut. In: M. Grötschel, (ed.), The Sharpest Cut, Festschrift in Honor of M. Padberg’s 60th Birthday. SIAM, to appearGoogle Scholar
32. 32.
Lovász, L.: On the Shannon capacity of a graph. IEEE Trans. Inform. Theory 25 (1), 1–7 (1979)Google Scholar
33. 33.
Lovász, L., Schrijver, A.: Cones of matrices and set-functions and 0-1 optimization. SIAM J. Optim. 1 (2), 166–190 (1991)Google Scholar
34. 34.
Nakata, K., Fujisawa, K., Fukuda, M., Kojima, M., Murota, K.: Exploiting sparsity in semidefinite programming via matrix completion II: Implementation and numerical results. Math. Prog. 95, 303–327 (2003)
35. 35.
Parrilo, P.A.: Semidefinite programming relaxations for semialgebraic problems. Math. Programming 96 (2, Ser. B), 293–320 (2003)Google Scholar
36. 36.
Sherali, H.D., Adams, W.P.: A hierarchy of relaxations between the continuous and convex hull representations for zero-one programming problems. SIAM J. Disc. Math. 3 (3), 411–430 (1990)
37. 37.
Sturm, J.F.: Using SeDuMi 1.02, a MATLAB toolbox for optimization over symmetric cones. Optim. Methods Softw. 11/12 (1–4), 625–653 (1999)Google Scholar
38. 38.
Toh, K.C., Todd, M.J., Tütüncü, R.H.: SDPT3 – a MATLAB software package for semidefinite programming, version 1. 3. Optim. Methods Softw. 11/12 (1–4), 545–581 (1999)Google Scholar
39. 39.
van Maaren, H.: Elliptic approximations of propositional formulae. Discrete Appl. Math. 96/97, 223–244 (1999)Google Scholar
40. 40.
Warners, J.P., van Maaren, H.: A two-phase algorithm for solving a class of hard satisfiability problems. Oper. Res. Lett. 23 (3–5), 81–88 (1998)Google Scholar
41. 41.
Wolkowicz, H., Saigal, R., vandenberghe, L., editors.: Handbook of semidefinite programming. Kluwer Academic Publishers, Boston, MA, 2000Google Scholar
42. 42.
Zwick, U.: Outward rotations: a tool for rounding solutions of semidefinite programming relaxations, with applications to MAX CUT and other problems. In: Proceedings of 31st STOC, 1999, pp. 679–687Google Scholar