RelEx: A Model-Agnostic Relational Model Explainer
This addresses the need for interpretability in relational AI models, which is crucial for domains like social networks or bioinformatics, but it is incremental as it builds on existing explainability work by extending it to relational contexts.
The paper tackles the problem of explaining relational models like Graph Neural Networks (GNNs) and Statistical Relational Learning (SRL), which are often black-box and not covered by existing iid-based explainability methods, by developing RelEx, a model-agnostic explainer that achieves comparable or better performance than state-of-the-art methods like GNN-Explainer.
In recent years, considerable progress has been made on improving the interpretability of machine learning models. This is essential, as complex deep learning models with millions of parameters produce state of the art results, but it can be nearly impossible to explain their predictions. While various explainability techniques have achieved impressive results, nearly all of them assume each data instance to be independent and identically distributed (iid). This excludes relational models, such as Statistical Relational Learning (SRL), and the recently popular Graph Neural Networks (GNNs), resulting in few options to explain them. While there does exist one work on explaining GNNs, GNN-Explainer, they assume access to the gradients of the model to learn explanations, which is restrictive in terms of its applicability across non-differentiable relational models and practicality. In this work, we develop RelEx, a model-agnostic relational explainer to explain black-box relational models with only access to the outputs of the black-box. RelEx is able to explain any relational model, including SRL models and GNNs. We compare RelEx to the state-of-the-art relational explainer, GNN-Explainer, and relational extensions of iid explanation models and show that RelEx achieves comparable or better performance, while remaining model-agnostic.