Counterfactual Explanations for Hypergraph Neural Networks

TL;DR

This paper introduces CF-HyperGNNExplainer, a method for generating minimal, actionable counterfactual explanations for hypergraph neural networks, enhancing interpretability by identifying key higher-order relations influencing predictions.

Contribution

The paper presents a novel counterfactual explanation approach specifically designed for hypergraph neural networks, focusing on minimal structural edits to explain model decisions.

Findings

CF-HyperGNNExplainer produces valid, concise counterfactuals.

The method highlights critical higher-order relations in HGNN decisions.

Experiments on benchmark datasets demonstrate effectiveness.

Abstract

Hypergraph neural networks (HGNNs) effectively model higher-order interactions in many real-world systems but remain difficult to interpret, limiting their deployment in high-stakes settings. We introduce CF-HyperGNNExplainer, a counterfactual explanation method for HGNNs that identifies the minimal structural changes required to alter a model's prediction. The method generates counterfactual hypergraphs using actionable edits limited to removing node-hyperedge incidences or deleting hyperedges, producing concise and structurally meaningful explanations. Experiments on three benchmark datasets show that CF-HyperGNNExplainer generates valid and concise counterfactuals, highlighting the higher-order relations most critical to HGNN decisions.