A General Framework for quantifying Aleatoric and Epistemic uncertainty in Graph Neural Networks

TL;DR

This paper introduces a unified Bayesian framework for quantifying both aleatoric and epistemic uncertainties in Graph Neural Networks, enhancing the interpretability of GNN predictions in real-world applications.

Contribution

It presents a novel approach combining Assumed Density Filtering and Monte Carlo dropout to jointly estimate data and model uncertainties in GNNs.

Findings

Bayesian GNNs perform comparably to frequentist models.

The framework effectively captures uncertainties from data and model.

Uncertainty estimates improve prediction reliability.

Abstract

Graph Neural Networks (GNN) provide a powerful framework that elegantly integrates Graph theory with Machine learning for modeling and analysis of networked data. We consider the problem of quantifying the uncertainty in predictions of GNN stemming from modeling errors and measurement uncertainty. We consider aleatoric uncertainty in the form of probabilistic links and noise in feature vector of nodes, while epistemic uncertainty is incorporated via a probability distribution over the model parameters. We propose a unified approach to treat both sources of uncertainty in a Bayesian framework, where Assumed Density Filtering is used to quantify aleatoric uncertainty and Monte Carlo dropout captures uncertainty in model parameters. Finally, the two sources of uncertainty are aggregated to estimate the total uncertainty in predictions of a GNN. Results in the real-world datasets demonstrate that the Bayesian model performs at par with a frequentist model and provides additional information about predictions uncertainty that are sensitive to uncertainties in the data and model.