Geometry of Semidefinite Max-Cut Relaxations via Matrix Ranks
Rent the article at a discountRent now
* Final gross prices may vary according to local VAT.Get Access
Semidefinite programming (SDP) relaxations are proving to be a powerful tool for finding tight bounds for hard discrete optimization problems. This is especially true for one of the easier NP-hard problems, the Max-Cut problem (MC). The well-known SDP relaxation for Max-Cut, here denoted SDP1, can be derived by a first lifting into matrix space and has been shown to be excellent both in theory and in practice.
Recently the present authors have derived a new relaxation using a second lifting. This new relaxation, denoted SDP2, is strictly tighter than the relaxation obtained by adding all the triangle inequalities to the well-known relaxation. In this paper we present new results that further describe the remarkable tightness of this new relaxation. Let \(F_n \) denote the feasible set of SDP2 for the complete graph with n nodes, let F n denote the appropriately defined projection of \(F_n \) into \(S^n \) , the space of real symmetric n × n matrices, and let C n denote the cut polytope in \(S^n \) . Further let \(Y \in F_n \) be the matrix variable of the SDP2 relaxation and X ∈ F n be its projection. Then for the complete graph on 3 nodes, F 3 = C 3 holds. We prove that the rank of the matrix variable \(Y \in F_3 \) of SDP2 completely characterizes the dimension of the face of the cut polytope in which the corresponding matrix X lies. This shows explicitly the connection between the rank of the variable Y of the second lifting and the possible locations of the projected matrix X within C 3. The results we prove for n = 3 cast further light on how SDP2 captures all the structure of C 3, and furthermore they are stepping stones for studying the general case n ≥ 4. For this case, we show that the characterization of the vertices of the cut polytope via rank Y = 1 extends to all n ≥ 4. More interestingly, we show that the characterization of the one-dimensional faces via rank Y = 2 also holds for n ≥ 4. Furthermore, we prove that if rank Y = 2 for n ≥ 3, then a simple algorithm exhibits the two rank-one matrices (corresponding to cuts) which are the vertices of the one-dimensional face of the cut polytope where X lies.
- M.F. Anjos, “New convex relaxations for the Maximum Cut and VLSI layout problems,” Ph.D. Thesis, University of Waterloo, 2001. Available online at http://etd.uwaterloo.ca/etd/manjos2001.pdf.
- M.F. Anjos and H. Wolkowicz, “Strengthened semidefinite relaxations via a second lifting for the Max-Cut problem,” Discrete Appl. Math., vol. 199, no. 1-2, pp. 79–106, 2002.
- F. Barahona, “The max-cut problem on graphs not contractible to K5,” Oper. Res. Lett., vol. 2, no. 3, pp. 107–111, 1983.
- F. Barahona, “On cuts and matchings in planar graphs,” Math. Programming, vol. 60, no. 1 (Ser. A), pp. 53–68, 1993.
- F. Barahona, M. Grötschel, M. J¨unger, and G. Reinelt, “An application of combinatorial optimization to statistical physics and circuit layout design,” Oper. Res., vol. 36, pp. 493–513, 1988.
- F. Barahona and A.R. Mahjoub, “On the cut polytope,” Math. Programming, vol. 36, no. 2, pp. 157–173, 1986.
- S.J. Benson, Y. Ye, and X. Zhang, “Solving large-scale sparse semidefinite programs for combinatorial optimization,” SIAM J. Optim., vol. 10, no. 2, pp. 443–461 (electronic), 2000.
- S. Burer and R.D.C. Monteiro, “A projected gradient algorithm for solving the maxcut SDP relaxation,” Optim. Meth. Softw., vol. 15, pp. 175–200, 2001.
- M.M. Deza and M. Laurent, Geometry of Cuts and Metrics, Springer-Verlag: Berlin, 1997.
- U. Feige and G. Schechtman, “On the optimality of the random hyper plane rounding technique for MAX CUT,” Technical report, Weizmann Institute, Rehovot, Israel, 2000.
- M.R. Garey and D.S. Johnson, Computers and Intractability: A Guide to the Theory of NP-Completeness, Freeman: San Francisco, 1979.
- M.X. Goemans, “Semi definite programming in combinatorial optimization,” Math. Programming, vol. 79, pp. 143–162, 1997.
- M.X. Goemans, “Semi definite programming and combinatorial optimization,” Documenta Mathematica, Extra Volume ICM 1998, pp. 657–666, 1998.
- M.X. Goemans and F. Rendl, “Combinatorial optimization,” in Handbook of Semidefinite Programming: Theory, Algorithms, and Applications, H. Wolkowicz, R. Saigal, and L. Vandenberghe (Eds.), Kluwer Academic Publishers: Boston, MA, 2000.
- M.X. Goemans and D.P. Williamson, “.878-approximation algorithms for MAX CUT and MAX 2SAT,” in ACM Symposium on Theory of Computing (STOC), 1994.
- M.X. Goemans and D.P. Williamson, “Improved approximation algorithms for maximum cut and satisfiability problems using semi definite programming,” J. Assoc. Comput. Mach., vol. 42, no. 6, pp. 1115–1145, 1995.
- J. Hastad, “Some optimal in approximability results,” in Proc. of the 29th ACM Symp. on Theory Comput., 1997.
- C. Helmberg, “An interior-point method for semi definite programming and max-cut bounds,” PhD Thesis, Graz University of Technology, Austria, 1994.
- C. Helmberg and F. Oustry, “Bundle methods to minimize the maximum eigenvalue function,” in Handbook of Semidefinite Programming: Theory, Algorithms, and Applications, H. Wolkowicz, R. Saigal, and L. Vandenberghe (Eds.), Kluwer Academic Publishers: Boston, MA, 2000.
- C. Helmberg, S. Poljak, F. Rendl, and H. Wolkowicz, “Combining semidefinite and polyhedral relaxations for integer programs,” in Integer Programming and Combinatorial Optimization (Copenhagen, 1995), Springer: Berlin, 1995, pp. 124–134.
- C. Helmberg and F. Rendl, “A spectral bundle method for semi definite programming,” SIAM J. Optim., vol. 10, no. 3, pp. 673–696, 2000.
- C. Helmberg, F. Rendl, R.J. Vanderbei, and H. Wolkowicz, “An interior-point method for semi definite programming,” SIAM J. Optim., vol. 6, no. 2, pp. 342–361, 1996.
- R.A. Horn and C.R. Johnson, Matrix Analysis, Cambridge University Press: Cambridge, 1990.
- R.M. Karp, “Reducibility among combinatorial problems,” in Complexity of Computer Computation, R.E. Miller and J.W. Thatcher (Eds.), Plenum Press: New York, 1972, pp. 85–103.
- J.B. Lasserre, “Optimality conditions and LMI relaxations for 0-1 programs,” LAAS research report 00099, LAAS-CNRS, Toulouse, France, 2000.
- M. Laurent, “Tighter linear and semi definite relaxations for max-cut based on the Lovász-Schrijver lift-and-project procedure,” SIAM J. Optim., vol. 12, pp. 345–375, 2001 (electronic publication).
- M. Laurent and S. Poljak, “On a positive semi definite relaxation of the cut polytope,” Linear Algebra Appl., vol. 223/224, pp. 439–461, 1995.
- M. Laurent and S. Poljak, “On the facial structure of the correlation matrices,” SIAM J. Matrix Anal. Appl., vol. 17, no. 3, pp. 530–547, 1996.
- T. Lengauer, Combinatorial Algorithms for Integrated Circuit Layout, JohnWiley & Sons Ltd.: Chichester, 1990.
- B. Mohar and S. Poljak, “Eigenvalues in combinatorial optimization,” in Combinatorial Graph-Theoretical Problems in Linear Algebra, IMA, vol. 50, Springer-Verlag: Berlin, 1993.
- Y.E. Nesterov, “Quality of semidefinite relaxation for non convex quadratic optimization,” Technical report, CORE, Universite Catholique de Louvain, Belgium, 1997.
- Y.E. Nesterov, H. Wolkowicz, and Y. Ye, “Semi definite programming relaxations of nonconvex quadratic optimization,” in Handbook of Semi definite Programming: Theory, Algorithms, and Applications, H. Wolkowicz, R. Saigal, and L. Vandenberghe (Eds.), Kluwer Academic Publishers: Boston, MA, 2000, p. 34.
- P. Parrilo, “Structured semi definite programs and semi algebraic geometry methods in robustness and optimization,” Ph.D. Thesis, Caltech, Pasadena, California, 2000.
- F. Rendl, “Semi definite programming and combinatorial optimization,” Appl. Numer. Math., vol. 29, pp. 255–281, 1999.
- H. Wolkowicz, R. Saigal, and L. Vandenberghe (Eds.), Handbook of Semi definite Programming: Theory, Algorithms, and Applications, Kluwer Academic Publishers: Boston, MA, 2000, pp. xxvi+654.
- Y. Ye, “Approximating quadratic programming with bound and quadratic constraints,” Math. Programming, vol. 84, pp. 219–226, 1999.
- Geometry of Semidefinite Max-Cut Relaxations via Matrix Ranks
Journal of Combinatorial Optimization
Volume 6, Issue 3 , pp 237-270
- Cover Date
- Print ISSN
- Online ISSN
- Kluwer Academic Publishers
- Additional Links
- semidefinite programming
- discrete optimization
- Lagrangian relaxation
- Max-Cut problem
- Industry Sectors