UNR-Explainer: Counterfactual Explanations for Unsupervised Node Representation Learning Models

2026-05-17

AI Summary

This work addresses the lack of effective interpretability methods in unsupervised node representation learning, particularly the absence of counterfactual explanations. It proposes the first counterfactual explanation framework tailored for unsupervised graph representation learning, generating concise and semantically meaningful explanations by identifying critical subgraphs that most influence the k-nearest neighbors of a target node in the embedding space. The approach efficiently explores the subgraph perturbation space by integrating Monte Carlo Tree Search (MCTS) with k-nearest neighbor sensitivity analysis. Experimental results demonstrate that the framework significantly improves explanation quality across models such as GraphSAGE and Deep Graph Infomax (DGI), effectively supporting downstream tasks including top-k link prediction and clustering.

Abstract

Node representation learning, such as Graph Neural Networks (GNNs), has emerged as a pivotal method in machine learning. The demand for reliable explanation generation surges, yet unsupervised models remain underexplored. To bridge this gap, we introduce a method for generating counterfactual (CF) explanations in unsupervised node representation learning. We identify the most important subgraphs that cause a significant change in the k-nearest neighbors of a node of interest in the learned embedding space upon perturbation. The k-nearest neighbor-based CF explanation method provides simple, yet pivotal, information for understanding unsupervised downstream tasks, such as top-k link prediction and clustering. Consequently, we introduce UNR-Explainer for generating expressive CF explanations for Unsupervised Node Representation learning methods based on a Monte Carlo Tree Search (MCTS). The proposed method demonstrates superior performance on diverse datasets for unsupervised GraphSAGE and DGI.

Problem

• counterfactual explanations
• unsupervised node representation learning
• graph neural networks
• k-nearest neighbors
• explainable AI

Innovation

• Counterfactual Explanation
• Unsupervised Node Representation Learning
• Monte Carlo Tree Search
• Graph Neural Networks
• Explainable AI