Layerwise Relevance Visualization in Convolutional Text Graph Classifiers
This work addresses interpretability challenges in deep learning for researchers and practitioners using GCNs, though it is incremental as it builds on existing explainability methods by focusing on intermediate states.
The authors tackled the problem of interpreting hidden layer representations in Graph Convolutional Networks (GCNs) by developing a method that traces and visualizes features contributing to classification decisions across layers, demonstrating meaningful layerwise explanations for a GCN sentence classifier.
Representations in the hidden layers of Deep Neural Networks (DNN) are often hard to interpret since it is difficult to project them into an interpretable domain. Graph Convolutional Networks (GCN) allow this projection, but existing explainability methods do not exploit this fact, i.e. do not focus their explanations on intermediate states. In this work, we present a novel method that traces and visualizes features that contribute to a classification decision in the visible and hidden layers of a GCN. Our method exposes hidden cross-layer dynamics in the input graph structure. We experimentally demonstrate that it yields meaningful layerwise explanations for a GCN sentence classifier.