Conformal Prediction Sets for Populations of Graphs

TL;DR

This paper introduces a conformal prediction methodology for graph data that provides valid uncertainty quantification for both labeled and unlabeled graphs without distributional assumptions, demonstrated through simulations and sports analysis.

Contribution

It develops a novel conformal prediction approach for graph populations, including unlabelled graphs, with finite-sample validity and interpretable prediction sets.

Findings

Method achieves finite-sample validity.

Applicable to both labeled and unlabeled graphs.

Demonstrated on FIFA 2018 football data.

Abstract

The analysis of data such as graphs has been gaining increasing attention in the past years. This is justified by the numerous applications in which they appear. Several methods are present to predict graphs, but much fewer to quantify the uncertainty of the prediction. The present work proposes an uncertainty quantification methodology for graphs, based on conformal prediction. The method works both for graphs with the same set of nodes (labelled graphs) and graphs with no clear correspondence between the set of nodes across the observed graphs (unlabelled graphs). The unlabelled case is dealt with the creation of prediction sets embedded in a quotient space. The proposed method does not rely on distributional assumptions, it achieves finite-sample validity, and it identifies interpretable prediction sets. To explore the features of this novel forecasting technique, we perform two simulation studies to show the methodology in both the labelled and the unlabelled case. We showcase the applicability of the method in analysing the performance of different teams during the FIFA 2018 football world championship via their player passing networks.