B1722
Title: A Bayesian approach to network classification
Authors: Sharmistha Guha - Texas A&M University (United States) [presenting]
Abstract: A novel Bayesian binary classification framework is proposed for networks with labelled nodes. The approach is motivated by applications in brain connectome studies, where the overarching goal is to identify both regions of interest in the brain and connections between ROIs that influence how study subjects are classified. A binary logistic regression framework is developed with the network as the predictor, and model the associated network coefficient using a novel class of global-local network shrinkage priors. A theoretical analysis of a member of this class of priors is performed, which is called the Network Lasso Prior, and shows the asymptotically correct classification of networks even when the number of network edges grows faster than the sample size. Two representative members from this class of priors, the Network Lasso prior and the Network Horseshoe prior, are implemented using an efficient Markov Chain Monte Carlo algorithm, and empirically evaluated through simulation studies and the analysis of a real brain connectome dataset.
