English

Efficient Clustering for Stretched Mixtures: Landscape and Optimality

Machine Learning 2021-11-30 v3 Machine Learning Optimization and Control Statistics Theory Methodology Statistics Theory

Abstract

This paper considers a canonical clustering problem where one receives unlabeled samples drawn from a balanced mixture of two elliptical distributions and aims for a classifier to estimate the labels. Many popular methods including PCA and k-means require individual components of the mixture to be somewhat spherical, and perform poorly when they are stretched. To overcome this issue, we propose a non-convex program seeking for an affine transform to turn the data into a one-dimensional point cloud concentrating around 1-1 and 11, after which clustering becomes easy. Our theoretical contributions are two-fold: (1) we show that the non-convex loss function exhibits desirable geometric properties when the sample size exceeds some constant multiple of the dimension, and (2) we leverage this to prove that an efficient first-order algorithm achieves near-optimal statistical precision without good initialization. We also propose a general methodology for clustering with flexible choices of feature transforms and loss objectives.

Keywords

Cite

@article{arxiv.2003.09960,
  title  = {Efficient Clustering for Stretched Mixtures: Landscape and Optimality},
  author = {Kaizheng Wang and Yuling Yan and Mateo Díaz},
  journal= {arXiv preprint arXiv:2003.09960},
  year   = {2021}
}

Comments

36 pages

R2 v1 2026-06-23T14:23:15.133Z