Finding Counterfactual Evidences for Node Classification

TL;DR

This paper introduces a method to identify counterfactual evidence in node classification tasks using GNNs, aiming to improve fairness and interpretability by analyzing similar nodes with different classifications.

Contribution

It proposes novel algorithms and indexing techniques to efficiently find counterfactual node pairs, generalizing across different GNN models and enhancing interpretability.

Findings

Effective algorithms for counterfactual evidence search

Enhanced fairness and accuracy in GNNs

Generalization across various GNN architectures

Abstract

Counterfactual learning is emerging as an important paradigm, rooted in causality, which promises to alleviate common issues of graph neural networks (GNNs), such as fairness and interpretability. However, as in many real-world application domains where conducting randomized controlled trials is impractical, one has to rely on available observational (factual) data to detect counterfactuals. In this paper, we introduce and tackle the problem of searching for counterfactual evidences for the GNN-based node classification task. A counterfactual evidence is a pair of nodes such that, regardless they exhibit great similarity both in the features and in their neighborhood subgraph structures, they are classified differently by the GNN. We develop effective and efficient search algorithms and a novel indexing solution that leverages both node features and structural information to identify counterfactual evidences, and generalizes beyond any specific GNN. Through various downstream applications, we demonstrate the potential of counterfactual evidences to enhance fairness and accuracy of GNNs.