Abstract
Euclidean distance embedding appears in many high-profile applications including wireless sensor network localization, where not all pairwise distances among sensors are known or accurate. The classical Multi-Dimensional Scaling (cMDS) generally works well when the partial or contaminated Euclidean Distance Matrix (EDM) is close to the true EDM, but otherwise performs poorly. A natural step preceding cMDS would be to calculate the nearest EDM to the known matrix. A crucial condition on the desired nearest EDM is for it to have a low embedding dimension and this makes the problem nonconvex. There exists a large body of publications that deal with this problem. Some try to solve the problem directly and some are the type of convex relaxations of it. In this paper, we propose a numerical method that aims to solve this problem directly. Our method is strongly motivated by the majorized penalty method of Gao and Sun for low-rank positive semi-definite matrix optimization problems. The basic geometric object in our study is the set of EDMs having a low embedding dimension. We establish a zero duality gap result between the problem and its Lagrangian dual problem, which also motivates the majorization approach adopted. Numerical results show that the method works well for the Euclidean embedding of Network coordinate systems and for a class of problems in large scale sensor network localization and molecular conformation.
Similar content being viewed by others
Notes
Available from http://pdos.lcs.mit.edu/p2psim/Kingdata.
Available from http://www.cs.cornell.edu/People/egs/meridian/data.php.
Available from http://www.cs.rice.edu/~eugeneng/research/gnp.
Data available from http://people.sc.fsu.edu/~jburkardt/datasets/cities/cities.html.
References
Alfakih, A.Y., Khandani, A., Wolkowicz, H.: Solving Euclidean distance matrix completion problems via semidefinite programming. Comput. Optim. Appl. 12, 13–30 (1999)
Berman, H.M., Westbrook, J., Feng, Z., Gillilan, G., Bhat, T.N., Weissig, H., Shindyalov, I.N., Bourne, P.E.: The protein data bank. Nucleic Acids Res. 28, 235–242 (2000)
Biswas, P., Ye, Y.: Semidefinite programming for ad hoc wireless sensor network localization. In: Proceedings of the 3rd IPSN, Berkeley, CA, pp. 46–54 (2004)
Biswas, P., Liang, T.-C., Toh, K.-C., Wang, T.-C., Ye, Y.: Semidefinite programming approaches for sensor network localization with noisy distance measurements. IEEE Trans. Autom. Sci. Eng. 3, 360–371 (2006)
Borg, I., Groenen, P.J.F.: Modern Multidimensional Scaling: Theory and Applications, Springer Series in Statistics, 2nd edn. Springer, Berlin (2005)
Cayton, L., Dasgupta, S.: Robust Euclidean embedding. In: Proceeding of the 23rd International Conference on Machine Learning, pp. 169–176, Pittsburgh, PA (2006)
Chu, D.I., Brown, H.C., Chu, M.T.: On least squares Euclidean distance matrix approximation and completion. Department of Mathematics, North Carolina State University, Tech. Rep. (2003)
Clarke, F.H.: Optimization and Nonsmooth Analysis. Wiley, New York (1983)
Cox, T.F., Cox, M.A.A.: Multidimensional Scaling, 2nd edn. Chapman and Hall/CRC, London (2001)
Crippen, G., Havel, T.: Distance Geometry and Molecular Conformation. Wiley, New York
Dattorro, J.: Convex Optimization and Euclidean Distance Geometry. Meboo Publishing, USA (2005)
de Leeuw, J.: An alternating least squares approach to squared distance scaling, unpublished manuscript, Department of Data Theory, University of Leiden, Leiden, The Netherlands (1975)
Dykstra, R.L.: An algorithm for restricted least squares regression. J. Am. Stat. Assoc. 78, 839–842 (1983)
Gaffke, N., Mathar, R.: A cyclic projection algorithm via duality. Metrika 36, 29–54 (1989)
Gao, Y., Sun, D.F.: A majorized penalty approach for calibrating rank constrained correlation matrix problems. Technical Report, Department of Mathematics, National University of Singapore, March (2010)
Gao, Y.: Structured Low Rank Matrix Optimization Problems: A Penalty Approach, PhD Thesis, National University of Singapore (2010)
Glunt, W., Hayden, T.L., Hong, S., Wells, J.: An alternating projection algorithm for computing the nearest Euclidean distance matrix. SIAM J. Matrix Anal. Appl. 11, 589–600 (1990)
Glunt, W., Hayden, T.L., Liu, W.-M.: The embedding problem for predistance matrices. Bull. Math. Biol. 53, 769–796 (1991)
Glunt, W., Hayden, T.L., Raydan, R.: Molecular conformations from distance matrices. J. Comput. Chem. 14, 114–120 (1993)
Gower, J.C.: Euclidean ditance geometry. Math. Sci. 7, 1–14 (1982)
Gower, J.C.: Properties of Euclidean and non-Euclidean distance matrices. Linear Algebra Appl. 67, 81–97 (1985)
Han, S.P.: A successive projection method. Math. Program. 40, 1–14 (1988)
Hayden, T.L., Wells, J.: Approximation by matrices positive semidefinite on a subspace. Linear Algebra Appl. 109, 115–130 (1988)
Hiriart-Urruty, J.-B., Lemaréchal, C.: Convex Analysis and Minimization Algorithms I. Springer, Berlin (1993)
Jiang, K.F., Sun, D., Toh, K.-C.: Solving nuclear norm regularized and semidefinite matrix least sqaures problems with linear quality constraints. In: Bezdek, K., Deze, A., Ye, Y. (eds.) Fields Institute Communications Series on Discrete Geometry and Optimization, pp. 133–162 (2013)
Jiang, K.F., Sun, D.F., Toh, K.-C.: An inexact accelerated proximal gradient method for large scale linearly constrained convex SDP. SIAM J. Optim. 22, 1042–1064 (2012)
Johnson, C.R., Tarazaga, P.: Connections between the real positive semidefinite and distance matrix completion problems. Linear Algebra Appl. 223(224), 375–391 (1995)
Kim, S., Kojima, M., Waki, H.: Exploiting sparsity in SDP relaxation for sensor network localization. SIAM J. Optim. 20, 192–215 (2009)
Krislock, N., Wolkowicz, H.: Euclidean distance matrices and applications. In: Anjos, M., Lasserre, J. (eds.) Handbook of Semidefinite, Cone and Polynomial Optimization, pp. 879–914 (2012)
Krislock, N., Wolkowicz, H.: Explicit sensor network localization using semidefinite representations and facial reductions. SIAM J. Optim. 20, 2679–2708 (2010)
Laurent, M.: A connection between positive semidefinite and Euclidean distance matrix completion problems. Linear Algebra Appl. 273, 9–22 (1998)
Laurent, M.: Matrix completion problem. In: Floudas, C.A., Pardalos, P.M. (eds.) The Encyclopedia of Optimization, vol. III, pp. 221–229. Kluwer, Dordrecht (2001)
Lavor, C., Liberti, L., Maculan, N., Mucherino, A.: Recent advances on the discretization molecular distance geometry problem. Eur. J. Oper. Res. 219, 698–706 (2012)
Lee, S., Zhang, Z., Sahu, S., Saha, D.: On suitability of Euclidean embedding for host-based network coordinate systems. IEEE/ACM Trans. Networking 18(1), 27–40 (2010)
Liberti, L., Lavor, C., Maculan, N., Mucherino, A.: Euclidean distance geometry and applications, available from arXiv:1205:0349v1. To appear in: SIAM Review May 3 (2012)
Liberti, L., Lavor, C., Mucherino, A., Maculan, N.: Molecular distance geometry methods: from continuous to discrete. Int. Trans. Oper. Res. 18, 33–51 (2011)
Mathar, R.: The best Euclidean fit to a given distance matrix in prescribed dimensions. Linear Algebra Appl. 67, 1–6 (1985)
Miao, W., Pan, S., Sun, D.: A rank-corrected procedure for matrix completion with fixed basis coefficients, Arxiv, preprint arXiv:1210.3709 (2012)
Miao, W.: Matrix completion procedure with fixed basis coefficients and rank regularized problems with hard constraints, PhD thesis, Department of Mathematics, National University of Singapore (2013)
Mishra, B., Meyer, G., Sepulchre, R.: Low-rank optimization for distance matrix completion. In: Proceedings of the 50th IEEE Conference on Decision and Control, December (2011)
Moré, J.J., Wu, Z.: Distance geometry optimization for protein structures. J. Glob. Optim. 15, 219–234 (1999)
Ng, T.E., Zhang, H.: Predicting Internet network distance with co-ordinates-based approaches, In: Proceedings of the IEEE INFOCOM, New York, pp. 170–179, June (2002)
Qi, H.-D.: A semismooth Newton method for the nearest Euclidean distance matrix problem. SIAM J. Matrix Anal. Appl. 34, 67–93 (2013)
Schoenberg, I.J.: Remarks to Maurice Fréchet’s article “Sur la définition axiomatque d’une classe d’espaces vectoriels distanciés applicbles vectoriellement sur l’espace de Hilbet”. Ann. Math. 36, 724–732 (1935)
Tenenbaum, J.B., de Silva, V., Langford, J.C.: A global geometric framework for nonlinear dimensionality reduction. Science 290, 2319–2323 (2000)
Toh, K.C.: An inexact path-following algorithm for convex quadratic SDP. Math. Program. 112, 221–254 (2008)
Trosset, M.W.: Distance matrix completion by numerical optimization. Comput. Optim. Appl. 17, 11–22 (2000)
Tseng, P.: Second-order cone programming relaxation of sensor network localization. SIAM J. Optim. 18, 156–185 (2007)
Young, G., Householder, A.S.: Discussion of a set of points in terms of their mutual distances. Psychometrika 3, 19–22 (1938)
Acknowledgments
The authors would like to thank Prof. Kim-Chuan Toh for his help on using the SDP solver as well as on the PDB problems. They also like to thank Bamdev Mishra for useful discussions on the manifold-based-optimization solver [40] (denoted by MBO solver in this paper). The authors also wish to thank the three referees as well as the AE for their valuable comments and constructive suggestions, which have greatly improved the quality and the presentation of the paper.
Author information
Authors and Affiliations
Corresponding author
Additional information
This research was supported by the EPSRC grant EP/K007645/1.
This author was supported by the General Research Fund from Hong Kong Research Grants Council: 203712.
Rights and permissions
About this article
Cite this article
Qi, HD., Yuan, X. Computing the nearest Euclidean distance matrix with low embedding dimensions. Math. Program. 147, 351–389 (2014). https://doi.org/10.1007/s10107-013-0726-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10107-013-0726-0
Keywords
- Euclidean distance matrix
- Lagrangian duality
- Low-rank approximation
- Majorization method
- Semismooth Newton-CG method