Convex Clustering via Optimal Mass Transport

Author(s): Carli, Francesca P; Ning, Lipeng; Georgiou, Tryphon T | Abstract: We consider approximating distributions within the framework of optimal mass transport and specialize to the problem of clustering data sets. Distances between distributions are measured in the Wasserstein metric. The main problem we consider is that of approximating sample distributions by ones with sparse support. This provides a new viewpoint to clustering. We propose different relaxations of a cardinality function which penalizes the size of the support set. We establish that a certain relaxation provides the tightest convex lower approximation to the cardinality penalty. We compare the performance of alternative relaxations on a numerical study on clustering.

[1]  A. Willsky,et al.  The Convex algebraic geometry of linear inverse problems , 2010, 2010 48th Annual Allerton Conference on Communication, Control, and Computing (Allerton).

[2]  Vijay V. Vazirani,et al.  Approximation algorithms for metric facility location and k-Median problems using the primal-dual schema and Lagrangian relaxation , 2001, JACM.

[3]  Stephen P. Boyd,et al.  A rank minimization heuristic with application to minimum order system approximation , 2001, Proceedings of the 2001 American Control Conference. (Cat. No.01CH37148).

[4]  Sudipto Guha,et al.  A constant-factor approximation algorithm for the k-median problem (extended abstract) , 1999, STOC '99.

[5]  Michael Elad,et al.  From Sparse Solutions of Systems of Equations to Sparse Modeling of Signals and Images , 2009, SIAM Rev..

[6]  Emmanuel J. Candès,et al.  Decoding by linear programming , 2005, IEEE Transactions on Information Theory.

[7]  M. Yuan,et al.  Model selection and estimation in regression with grouped variables , 2006 .

[8]  Stephen P. Boyd,et al.  Convex Optimization , 2004, Algorithms and Theory of Computation Handbook.

[9]  Polina Golland,et al.  Convex Clustering with Exemplar-Based Models , 2007, NIPS.

[10]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[11]  Mark S. Daskin,et al.  Network and Discrete Location: Models, Algorithms and Applications , 1995 .

[12]  Venkat Chandrasekaran,et al.  Recovery of Sparse Probability Measures via Convex Programming , 2012, NIPS.

[13]  Zvi Drezner,et al.  Facility location - applications and theory , 2001 .

[14]  Jeffrey Scott Vitter,et al.  e-approximations with minimum packing constraint violation (extended abstract) , 1992, STOC '92.

[15]  Arthur Cayley,et al.  The Collected Mathematical Papers: On Monge's “Mémoire sur la théorie des déblais et des remblais” , 2009 .

[16]  J. Vitter,et al.  Approximations with Minimum Packing Constraint Violation , 1992 .

[17]  C. Villani.  Optimal Transport: Old and New , 2008 .

[18]  Michael A. Saunders,et al.  Atomic Decomposition by Basis Pursuit , 1998, SIAM J. Sci. Comput..