GraphPPD: Posterior Predictive Modelling for Graph-Level Inference

TL;DR

This paper introduces GraphPPD, a variational framework for modeling the posterior predictive distribution in graph-level tasks, enabling uncertainty-aware predictions with improved effectiveness demonstrated on benchmark datasets.

Contribution

It presents a novel variational modeling framework for the posterior predictive distribution tailored for graph-level learning, addressing limitations of existing uncertainty methods.

Findings

Effective uncertainty modeling in graph-level tasks

Improved prediction safety and confidence understanding

Validated on multiple benchmark datasets

Abstract

Accurate modelling and quantification of predictive uncertainty is crucial in deep learning since it allows a model to make safer decisions when the data is ambiguous and facilitates the users' understanding of the model's confidence in its predictions. Along with the tremendously increasing research focus on \emph{graph neural networks} (GNNs) in recent years, there have been numerous techniques which strive to capture the uncertainty in their predictions. However, most of these approaches are specifically designed for node or link-level tasks and cannot be directly applied to graph-level learning problems. In this paper, we propose a novel variational modelling framework for the \emph{posterior predictive distribution}~(PPD) to obtain uncertainty-aware prediction in graph-level learning tasks. Based on a graph-level embedding derived from one of the existing GNNs, our framework can learn the PPD in a data-adaptive fashion. Experimental results on several benchmark datasets exhibit the effectiveness of our approach.