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

TL;DR

This paper introduces a scalable nonparametric Bayesian method for binary classification on large graphs, utilizing truncated Laplacian eigenfunctions to improve computational efficiency and accuracy.

Contribution

The paper proposes a truncated Laplacian eigenfunction prior for Bayesian graph classification, enhancing scalability over traditional untruncated methods.

Findings

Improved scalability on large graphs

Better classification accuracy in experiments

Effective comparison with untruncated prior

Abstract

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