Evaluating Link Prediction Explanations for Graph Neural Networks
This work addresses the lack of validation for link prediction explanations, which is crucial for fostering adoption of Graph Machine Learning in real-world domains, but it is incremental as it builds on existing explainability methods.
The paper tackles the problem of validating explanations for link prediction in Graph Neural Networks by introducing quantitative metrics to assess explanation quality, with or without ground-truth, and evaluates state-of-the-art explainability methods using these metrics.
Graph Machine Learning (GML) has numerous applications, such as node/graph classification and link prediction, in real-world domains. Providing human-understandable explanations for GML models is a challenging yet fundamental task to foster their adoption, but validating explanations for link prediction models has received little attention. In this paper, we provide quantitative metrics to assess the quality of link prediction explanations, with or without ground-truth. State-of-the-art explainability methods for Graph Neural Networks are evaluated using these metrics. We discuss how underlying assumptions and technical details specific to the link prediction task, such as the choice of distance between node embeddings, can influence the quality of the explanations.