Uncertainty in Graph Contrastive Learning with Bayesian Neural Networks

TL;DR

This paper introduces a Bayesian neural network approach to graph contrastive learning, enhancing uncertainty estimation and improving semi-supervised node classification performance on large unlabeled datasets.

Contribution

It proposes a novel Bayesian method for graph contrastive learning that improves uncertainty quantification and downstream task accuracy.

Findings

Enhanced uncertainty estimates with Bayesian neural networks.

Improved semi-supervised node classification accuracy.

A new measure of uncertainty based on likelihood disagreement.

Abstract

Graph contrastive learning has shown great promise when labeled data is scarce, but large unlabeled datasets are available. However, it often does not take uncertainty estimation into account. We show that a variational Bayesian neural network approach can be used to improve not only the uncertainty estimates but also the downstream performance on semi-supervised node-classification tasks. Moreover, we propose a new measure of uncertainty for contrastive learning, that is based on the disagreement in likelihood due to different positive samples.