GraphPPD: Posterior Predictive Modelling for Graph-Level Inference

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pal, Soumyasundar, Ma, Liheng, Natik, Amine, Zhang, Yingxue, Coates, Mark
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914001753997312
author Pal, Soumyasundar
Ma, Liheng
Natik, Amine
Zhang, Yingxue
Coates, Mark
author_facet Pal, Soumyasundar
Ma, Liheng
Natik, Amine
Zhang, Yingxue
Coates, Mark
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2508_16995
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle GraphPPD: Posterior Predictive Modelling for Graph-Level Inference
Pal, Soumyasundar
Ma, Liheng
Natik, Amine
Zhang, Yingxue
Coates, Mark
Machine Learning
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.
title GraphPPD: Posterior Predictive Modelling for Graph-Level Inference
topic Machine Learning
url https://arxiv.org/abs/2508.16995