On spectral clustering under non-isotropic Gaussian mixture models
Statistics Theory
2026-04-13 v2 Statistics Theory
Abstract
We evaluate the misclustering probability of a spectral clustering algorithm under a Gaussian mixture model with a general covariance structure. The algorithm partitions the data into two groups based on the sign of the first principal component score. As a corollary of the main result, the clustering procedure is shown to be consistent in a high-dimensional regime.
Cite
@article{arxiv.2601.13930,
title = {On spectral clustering under non-isotropic Gaussian mixture models},
author = {Kohei Kawamoto and Yuichi Goto and Koji Tsukuda},
journal= {arXiv preprint arXiv:2601.13930},
year = {2026}
}
Comments
8 pages