Spectral Detection in the Censored Block Model
Social and Information Networks
2020-01-22 v2 Disordered Systems and Neural Networks
Machine Learning
Probability
Abstract
We consider the problem of partially recovering hidden binary variables from the observation of (few) censored edge weights, a problem with applications in community detection, correlation clustering and synchronization. We describe two spectral algorithms for this task based on the non-backtracking and the Bethe Hessian operators. These algorithms are shown to be asymptotically optimal for the partial recovery problem, in that they detect the hidden assignment as soon as it is information theoretically possible to do so.
Cite
@article{arxiv.1502.00163,
title = {Spectral Detection in the Censored Block Model},
author = {Alaa Saade and Florent Krzakala and Marc Lelarge and Lenka Zdeborová},
journal= {arXiv preprint arXiv:1502.00163},
year = {2020}
}
Comments
ISIT 2015