English

Community Detection with Contextual Multilayer Networks

Statistics Theory 2023-01-13 v3 Information Theory math.IT Statistics Theory

Abstract

In this paper, we study community detection when we observe mm sparse networks and a high dimensional covariate matrix, all encoding the same community structure among nn subjects. In the asymptotic regime where the number of features pp and the number of subjects nn grows proportionally, we derive an exact formula of asymptotic minimum mean square error (MMSE) for estimating the common community structure in the balanced two block case. The formula implies the necessity of integrating information from multiple data sources. Consequently, it induces a sharp threshold of phase transition between the regime where detection (i.e., weak recovery) is possible and the regime where no procedure performs better than a random guess. The asymptotic MMSE depends on the covariate signal-to-noise ratio in a more subtle way than the phase transition threshold does. In the special case of m=1m=1, our asymptotic MMSE formula complements the pioneering work of Deshpande et. al. (2018) which found the sharp threshold when m=1m=1.

Keywords

Cite

@article{arxiv.2104.02960,
  title  = {Community Detection with Contextual Multilayer Networks},
  author = {Zongming Ma and Sagnik Nandy},
  journal= {arXiv preprint arXiv:2104.02960},
  year   = {2023}
}

Comments

76 pages, 2 figures

R2 v1 2026-06-24T00:54:50.939Z