English

Optimal Estimation of the Number of Communities

Methodology 2022-01-27 v2 Statistics Theory Statistics Theory

Abstract

In network analysis, how to estimate the number of communities KK is a fundamental problem. We consider a broad setting where we allow severe degree heterogeneity and a wide range of sparsity levels, and propose Stepwise Goodness-of-Fit (StGoF) as a new approach. This is a stepwise algorithm, where for m=1,2,m = 1, 2, \ldots, we alternately use a community detection step and a goodness-of-fit (GoF) step. We adapt SCORE \cite{SCORE} for community detection, and propose a new GoF metric. We show that at step mm, the GoF metric diverges to \infty in probability for all m<Km < K and converges to N(0,1)N(0,1) if m=Km = K. This gives rise to a consistent estimate for KK. Also, we discover the right way to define the signal-to-noise ratio (SNR) for our problem and show that consistent estimates for KK do not exist if SNR\goto0\mathrm{SNR} \goto 0, and StGoF is uniformly consistent for KK if SNR\goto\mathrm{SNR} \goto \infty. Therefore, StGoF achieves the optimal phase transition. Similar stepwise methods (e.g., \cite{wang2017likelihood, ma2018determining}) are known to face analytical challenges. We overcome the challenges by using a different stepwise scheme in StGoF and by deriving sharp results that are not available before. The key to our analysis is to show that SCORE has the {\it Non-Splitting Property (NSP)}. Primarily due to a non-tractable rotation of eigenvectors dictated by the Davis-Kahan sin(θ)\sin(\theta) theorem, the NSP is non-trivial to prove and requires new techniques we develop.

Keywords

Cite

@article{arxiv.2009.09177,
  title  = {Optimal Estimation of the Number of Communities},
  author = {Jiashun Jin and Zheng Tracy Ke and Shengming Luo and Minzhe Wang},
  journal= {arXiv preprint arXiv:2009.09177},
  year   = {2022}
}