Exploring Explainability Methods for Graph Neural Networks
This work addresses the need for explainability in graph neural networks, particularly for real-world applications, but is incremental as it tests known methods on a specific model.
The paper applied existing explainability methods to Graph Attention Networks for a super-pixel image classification task, evaluating their performance on three datasets to provide new insights into explainability in GNNs.
With the growing use of deep learning methods, particularly graph neural networks, which encode intricate interconnectedness information, for a variety of real tasks, there is a necessity for explainability in such settings. In this paper, we demonstrate the applicability of popular explainability approaches on Graph Attention Networks (GAT) for a graph-based super-pixel image classification task. We assess the qualitative and quantitative performance of these techniques on three different datasets and describe our findings. The results shed a fresh light on the notion of explainability in GNNs, particularly GATs.