Technical Program

Paper Detail

Paper Title Mutual Information for the Stochastic Block Model by the Adaptive Interpolation Method
Paper IdentifierMO3.R5.2
Authors Jean Barbier, Ecole Normale Supérieure Paris, France; Chun Lam Chan, Nicolas Macris, École polytechnique fédérale de Lausanne (EPFL), Switzerland
Session Community Detection and Graphical Models
Location Saint Victor, Level 3
Session Time Monday, 08 July, 14:30 - 16:10
Presentation Time Monday, 08 July, 14:50 - 15:10
Manuscript  Click here to download the manuscript
Abstract We rigorously derive a single-letter variational expression for the mutual information of the asymmetric two-groups stochastic block model in the dense graph regime. Existing proofs in the literature are indirect, as they involve mapping the model to a rank-one matrix estimation problem whose mutual information is then determined by a combination of methods (e.g., interpolation, cavity, algorithmic, spatial coupling). In this contribution we provide a self-contained direct method using only the recently introduced adaptive interpolation method.