A Tractable Fully Bayesian Method for the Stochastic Block Model
arXiv:1602.02256
Abstract
The stochastic block model (SBM) is a generative model revealing macroscopic structures in graphs. Bayesian methods are used for (i) cluster assignment inference and (ii) model selection for the number of clusters. In this paper, we study the behavior of Bayesian inference in the SBM in the large sample limit. Combining variational approximation and Laplace's method, a consistent criterion of the fully marginalized log-likelihood is established. Based on that, we derive a tractable algorithm that solves tasks (i) and (ii) concurrently, obviating the need for an outer loop to check all model candidates. Our empirical and theoretical results demonstrate that our method is scalable in computation, accurate in approximation, and concise in model selection.
References in corpus (16)
- Community structure in social and biological networks
- Lethality and centrality in protein networks
- Finding community structure in networks using the eigenvectors of matrices
- Self-similar community structure in organisations
- Asymptotic analysis of the stochastic block model for modular networks and its algorithmic applications
- Spectral redemption: clustering sparse networks
- Mixture models and exploratory analysis in networks
- A Bayesian Approach to Network Modularity
- Parsimonious module inference in large networks
- Uncovering latent structure in valued graphs: A variational approach
- Community Detection as an Inference Problem
- Entropy of stochastic blockmodel ensembles
- Model selection and hypothesis testing for large-scale network models with overlapping groups
- A simple example of Dirichlet process mixture inconsistency for the number of components
- (Un)detectable cluster structure in sparse networks
- Limitations in the spectral method for graph partitioning: detectability threshold and localization of eigenvectors