
How Robust Are Graph Neural Networks to Structural Noise?
Graph neural networks (GNNs) are an emerging model for learning graph em...
read it

GNN Explainer: A Tool for Posthoc Explanation of Graph Neural Networks
Graph Neural Networks (GNNs) are a powerful tool for machine learning on...
read it

XAI for Graphs: Explaining Graph Neural Network Predictions by Identifying Relevant Walks
Graph Neural Networks (GNNs) are a popular approach for predicting graph...
read it

Discovering Symbolic Models from Deep Learning with Inductive Biases
We develop a general approach to distill symbolic representations of a l...
read it

Learning to Drop: Robust Graph Neural Network via Topological Denoising
Graph Neural Networks (GNNs) have shown to be powerful tools for graph a...
read it

Graph Neural Network for Interpreting TaskfMRI Biomarkers
Finding the biomarkers associated with ASD is helpful for understanding ...
read it

Graph Ensemble Learning over Multiple Dependency Trees for Aspectlevel Sentiment Classification
Recent work on aspectlevel sentiment classification has demonstrated th...
read it
Interpreting Graph Neural Networks for NLP With Differentiable Edge Masking
Graph neural networks (GNNs) have become a popular approach to integrating structural inductive biases into NLP models. However, there has been little work on interpreting them, and specifically on understanding which parts of the graphs (e.g. syntactic trees or coreference structures) contribute to a prediction. In this work, we introduce a posthoc method for interpreting the predictions of GNNs which identifies unnecessary edges. Given a trained GNN model, we learn a simple classifier that, for every edge in every layer, predicts if that edge can be dropped. We demonstrate that such a classifier can be trained in a fully differentiable fashion, employing stochastic gates and encouraging sparsity through the expected L_0 norm. We use our technique as an attribution method to analyze GNN models for two tasks – question answering and semantic role labeling – providing insights into the information flow in these models. We show that we can drop a large proportion of edges without deteriorating the performance of the model, while we can analyse the remaining edges for interpreting model predictions.
READ FULL TEXT
Comments
There are no comments yet.