Abstract
We study semidefinite programming (SDP) relaxations for the NP-hard problem of globally optimizing a quadratic function over the Stiefel manifold. We introduce a strengthened relaxation based on two recent ideas in the literature: (i) a tailored SDP for objectives with a block-diagonal Hessian; and (ii) the use of the Kronecker matrix product to construct SDP relaxations. Using synthetic instances on five problem classes, we show that, in general, our relaxation significantly strengthens existing relaxations, although at the expense of longer solution times.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10957-023-02168-6/MediaObjects/10957_2023_2168_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10957-023-02168-6/MediaObjects/10957_2023_2168_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10957-023-02168-6/MediaObjects/10957_2023_2168_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10957-023-02168-6/MediaObjects/10957_2023_2168_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10957-023-02168-6/MediaObjects/10957_2023_2168_Fig5_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10957-023-02168-6/MediaObjects/10957_2023_2168_Fig6_HTML.png)
Similar content being viewed by others
Notes
This result and proof was shared with us by Dr. Gang Luo of Huawai Technologies Compnay, whom we gratefully acknowledge.
When \(\text {rank}(Y) > 1\), one could also try to reduce the rank of Y by moving to an extreme point of the optimal solution set using a procedure such as the rank-reduction algorithm described in [29]. If the subsequent rank is 1, then one recovers a primal value as just discussed. However, we have not implemented such a rank-reduction procedure in our computational results.
References
Absil, P.-A., Mahony, R., Sepulchre, R.: Optimization Algorithms on Matrix Manifolds. Princeton University Press, Princeton, NJ (2008)
Anjos, M. F., Wolkowicz, H.: Strengthened semidefinite relaxations via a second lifting for the Max-Cut problem. volume 119, pages 79–106. Foundations of heuristics in combinatorial optimization ( 2002)
Anstreicher, K., Chen, X., Wolkowicz, H., Yuan, Y.-X.: Strong duality for a trust-region type relaxation of the quadratic assignment problem. Linear Algebra Appl. 301(1–3), 121–136 (1999)
Anstreicher, K., Wolkowicz, H.: On Lagrangian relaxation of quadratic matrix constraints. SIAM J. Matrix Anal. Appl. 22(1), 41–55 (2000)
Anstreicher, K.M.: Kronecker product constraints with an application to the two-trust-region subproblem. SIAM J. Opt. 27(1), 368–378 (2017)
Birtea, P., Caşu, I., Comănescu, D.: Second order optimality on orthogonal Stiefel manifolds. Bulletin des Sci. Mathématiques 161, 102868 (2020)
Boumal, N.: An introduction to optimization on smooth manifolds. To appear with Cambridge University Press, (Apr 2022)
Boumal, N., Mishra, B., Absil, P.-A., Sepulchre, R.: Manopt, a Matlab toolbox for optimization on manifolds. J. Mach. Learn. Res. 15(42), 1455–1459 (2014)
Breloy, A., Kumar, S., Sun, Y., Palomar, D.P.: Majorization-minimization on the Stiefel manifold with application to robust sparse PCA. IEEE Transact. Signal Process. 69, 1507–1520 (2021)
Chrétien, S., Guedj, B.: Revisiting clustering as matrix factorisation on the Stiefel manifold. In G. Nicosia, V. Ojha, E. La Malfa, G. Jansen, V. Sciacca, P. Pardalos, G. Giuffrida, and R. Umeton, editors, International Conference on Machine Learning, Optimization, and Data Science, pp. 1–12. Springer, (2020)
Chu, M.T., Trendafilov, N.T.: The orthogonally constrained regression revisited. J. Comput. Graph. Stat. 10(4), 746–771 (2001)
de Klerk, E., Sotirov, R.: Exploiting group symmetry in semidefinite programming relaxations of the quadratic assignment problem. Math. Program 122(2), 225–246 (2010)
de Meijer, F., Sotirov, R., Wiegele, A., Zhao, S.: Partitioning through projections: strong sdp bounds for large graph partition problems, (2022)
Dodig, M., Stošić, M., Xavier, J.: On minimizing a quadratic function on Stiefel manifold. Linear Algebra Appl. 475, 251–264 (2015)
Edelman, A., Arias, T.A., Smith, S.T.: The geometry of algorithms with orthogonality constraints. SIAM J. Matrix Anal. Appl. 20(2), 303–353 (1998)
Eldén, L.: Algorithms for the regularization of ill-conditioned least squares problems. BIT Num. Math. 17(2), 134–145 (1977)
Eldén, L., Park, H.: A Procrustes problem on the Stiefel manifold. Num. Math. 82(4), 599–619 (1999)
Fu, M., Luo, Z.-Q., Ye, Y.: Approximation algorithms for quadratic programming. J. Combinat. Opt. 2, 29–50 (1998)
Gilman, K., Burer, S., Balzano, L.: A semidefinite relaxation for sums of heterogeneous quadratics on the Stiefel manifold, (2022)
Gower, J. C., Dijksterhuis, G. B.: Procrustes Problems. Oxford University Press, (2004)
Hu, J., Liu, X., Wen, Z.-W., Yuan, Y.-X.: A brief introduction to manifold optimization. J. Operat. Res. Soc. China 8(2), 199–248 (2020)
Jiang, R., Li, D.: Second order cone constrained convex relaxations for nonconvex quadratically constrained quadratic programming. J. Global Opt. 75(2), 461–494 (2019)
Kim, J., Kang, M., Kim, D., Ha, S.-Y., Yang, I.: A stochastic consensus method for nonconvex optimization on the Stiefel manifold. In 2020 59th IEEE Conference on Decision and Control (CDC), pp. 1050–1057. IEEE, (2020)
Lasserre, J.B.: Global optimization with polynomials and the problem of moments. SIAM J. Optim. 11(3), 796–817 (2000)
Li, X., Pong, T.K., Sun, H., Wolkowicz, H.: A strictly contractive Peaceman-Rachford splitting method for the doubly nonnegative relaxation of the minimum cut problem. Comput. Opt. Appl. 78(3), 853–891 (2021)
Nemirovski, A.: Sums of random symmetric matrices and quadratic optimization under orthogonality constraints. Math Program. 109(2), 283–317 (2007)
Nesterov, Y., Wolkowicz, H., Ye, Y.: Semidefinite programming relaxations of nonconvex quadratic optimization. In R. Saigal, L. Vandenberghe, and H. Wolkowicz, editors, Handbook of semidefinite programming, vol 27 of Internat. Ser. Oper. Res. Management Sci., pages 361–419. Kluwer Acad. Publ., Boston, MA, (2000)
Overton, M.L., Womersley, R.S.: On the sum of the largest eigenvalues of a symmetric matrix. SIAM J. Matrix Anal. Appl. 13(1), 41–45 (1992)
Pataki, G.: The geometry of semidefinite programming. In H. Wolkowicz, R. Saigal, and L. Vandenberghe, editors, Handbook of Semidefinite Programming: Theory, Algorithms, and Applications. Kluwer Academic Publishers, (2000)
Sanyal, R., Sottile, F., Sturmfels, B.: Orbitopes. Mathematika 57(2), 275–314 (2011)
Saunderson, J., Parrilo, P.A., Willsky, A.S.: Semidefinite descriptions of the convex hull of rotation matrices. SIAM J. Optim. 25(3), 1314–1343 (2015)
Sherali, H. D., Adams, W. P.: A Reformulation-Linearization Technique for Solving Discrete and Continuous Nonconvex Problems. Kluwer, (1997)
Smith, S. T.: Geometric Optimization Methods for Adaptive Filtering. PhD thesis, Harvard University, Cambridge, MA, (May 1993)
Stern, R.J., Wolkowicz, H.: Indefinite trust region subproblems and nonsymmetric eigenvalue perturbations. SIAM J. Optim. 5(2), 286–313 (1995)
Wen, Z., Yin, W.: A feasible method for optimization with orthogonality constraints. Math. Program. 142(1), 397–434 (2013)
Wolkowicz, H., Zhao, Q.: Semidefinite programming relaxations for the graph partitioning problem. vol 96/97, pages 461–479. 1999. The satisfiability problem (Certosa di Pontignano, 1996); Boolean functions
**a, Y., Han, Y.-W.: Partial Lagrangian relaxation for the unbalanced orthogonal Procrustes problem. Math. Methods Oper. Res. 79(2), 225–237 (2014)
Ye, Y.: A new complexity result on minimization of a quadratic function with a sphere constraint. In: Floudas, C., Pardalos, P. (eds.) Recent Advances in Global Optimization. Princeton University Press, Princeton, NJ (1992)
Zhang, X., Zhu, J., Wen, Z., Zhou, A.: Gradient type optimization methods for electronic structure calculations. SIAM J. Sci. Comput. 36(3), C265–C289 (2014)
Zhao, Q., Karisch, S. E., Rendl, F., Wolkowicz, H.: Semidefinite programming relaxations for the quadratic assignment problem. volume 2, pages 71–109. 1998. Semidefinite programming and interior-point approaches for combinatorial optimization problems (Toronto, ON, 1996)
Zohrizadeh, F., Kheirandishfard, M., Kamangar, F., Madani, R.: Non-smooth optimization over stiefel manifolds with applications to dimensionality reduction and graph clustering. In Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI-19, pp 1319–1326. International Joint Conferences on Artificial Intelligence Organization, 7 (2019)
Acknowledgements
We thank Kurt Anstreicher and Laura Balzano for contributing extremely helpful comments on a draft version of this research. In addition, we appreciate the constructive comments and suggestions of the two anonymous referees and handling editor. Finally, we express our thanks to Yong **a for providing a critical reference and Gang Luo for sharing Proposition 2.1 with us.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Etienne de Klerk.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Burer, S., Park, K. A Strengthened SDP Relaxation for Quadratic Optimization Over the Stiefel Manifold. J Optim Theory Appl (2023). https://doi.org/10.1007/s10957-023-02168-6
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10957-023-02168-6