Score: 1

An Observation on Lloyd's k-Means Algorithm in High Dimensions

Published: June 17, 2025 | arXiv ID: 2506.14952v1

By: David Silva-Sánchez, Roy R. Lederman

Potential Business Impact:

Fixes computer grouping when data is messy.

Business Areas:
Big Data Data and Analytics

Clustering and estimating cluster means are core problems in statistics and machine learning, with k-means and Expectation Maximization (EM) being two widely used algorithms. In this work, we provide a theoretical explanation for the failure of k-means in high-dimensional settings with high noise and limited sample sizes, using a simple Gaussian Mixture Model (GMM). We identify regimes where, with high probability, almost every partition of the data becomes a fixed point of the k-means algorithm. This study is motivated by challenges in the analysis of more complex cases, such as masked GMMs, and those arising from applications in Cryo-Electron Microscopy.

Country of Origin
🇺🇸 United States

Repos / Data Links

Page Count
27 pages

Category
Statistics:
Machine Learning (Stat)