Inference for the stochastic block model with unknown number of blocks and non-conjugate edge models
share
The stochastic block model (SBM) is a popular model for capturing community structure and interaction within a network. There is increasingly a range of network data with non-binary edges becoming commonplace. However, existing methods for inferring parameters of an SBM are commonly restricted to models for edges which yield a conjugate prior distribution. This paper introduces an effective reversible jump Markov chain Monte Carlo sampler for estimating the parameters and the number of blocks for a general network model allowing for a non-conjugate prior distribution on the edge model.
READ FULL TEXT
