Recovery guarantees for exemplar-based clustering

Abstract

For a certain class of distributions, we prove that the linear programming relaxation of $k$-medoids clustering---a variant of $k$-means clustering where means are replaced by exemplars from within the dataset---distinguishes points drawn from nonoverlapping balls with high probability once the number of points drawn and the separation distance between any two balls are sufficiently large. Our results hold in the nontrivial regime where the separation distance is small enough that points drawn from different balls may be closer to each other than points drawn from the same ball; in this case, clustering by thresholding pairwise distances between points can fail. We also exhibit numerical evidence of high-probability recovery in a substantially more permissive regime.

24 pages, 4 figures

References in corpus (8)

• Polynomial Learning of Distribution Families
• Clustering by soft-constraint affinity propagation: Applications to gene-expression data
• Random Projections for $k$-means Clustering
• Guaranteed clustering and biclustering via semidefinite programming
• Finding Dense Clusters via "Low Rank + Sparse" Decomposition
• Clustering using Max-norm Constrained Optimization
• Sparse Subspace Clustering: Algorithm, Theory, and Applications
• A Constant Factor Approximation Algorithm for Fault-Tolerant k-Median

Cited by in corpus (1)

• Relax, no need to round: integrality of clustering formulations