English

Eigen selection in spectral clustering: a theory guided practice

Statistics Theory 2020-06-09 v2 Statistics Theory

Abstract

Based on a Gaussian mixture type model , we derive an eigen selection procedure that improves the usual spectral clustering in high-dimensional settings. Concretely, we derive the asymptotic expansion of the spiked eigenvalues under eigenvalue multiplicity and eigenvalue ratio concentration results, giving rise to the first theory-backed eigen selection procedure in spectral clustering. The resulting eigen-selected spectral clustering (ESSC) algorithm enjoys better stability and compares favorably against canonical alternatives. We demonstrate the advantages of ESSC using extensive simulation and multiple real data studies.

Keywords

Cite

@article{arxiv.2004.06296,
  title  = {Eigen selection in spectral clustering: a theory guided practice},
  author = {Xiao Han and Xin Tong and Yingying Fan},
  journal= {arXiv preprint arXiv:2004.06296},
  year   = {2020}
}
R2 v1 2026-06-23T14:50:15.918Z