Nonparametric Bayesian label prediction on a large graph using truncated Laplacian regularization

This article describes an implementation of a nonparametric Bayesian approach\nto solving binary classification problems on graphs. We consider a hierarchical\nBayesian approach with a prior that is constructed by truncating a series\nexpansion of the soft label function using the graph Laplacian eigenfunctions\nas basis functions. We compare our truncated prior to the untruncated Laplacian\nbased prior in simulated and real data examples to illustrate the improved\nscalability in terms of size of the underlying graph.\n