Visualizing Graph Neural Networks with CorGIE: Corresponding a Graph to Its Embedding

Abstract

Graph neural networks (GNNs) are a class of powerful machine learning tools that model node relations for making predictions of nodes or links. GNN developers rely on quantitative metrics of the predictions to evaluate a GNN, but similar to many other neural networks, it is difficult for them to understand if the GNN truly learns characteristics of a graph as expected. We propose an approach to corresponding an input graph to its node embedding (aka latent space), a common component of GNNs that is later used for prediction. We abstract the data and tasks, and develop an interactive multi-view interface called CorGIE to instantiate the abstraction. As the key function in CorGIE, we propose the K-hop graph layout to show topological neighbors in hops and their clustering structure. To evaluate the functionality and usability of CorGIE, we present how to use CorGIE in two usage scenarios, and conduct a case study with five GNN experts.

References in corpus (5)

• Semi-Supervised Classification with Graph Convolutional Networks
• LINE: Large-scale Information Network Embedding
• The State of the Art in Enhancing Trust in Machine Learning Models with the Use of Visualizations
• Embedding Projector: Interactive Visualization and Interpretation of Embeddings
• Explainability in Graph Neural Networks: A Taxonomic Survey

Cited by in corpus (2)

• Grounding force-directed network layouts with latent space models
• InteractiveGNNExplainer: A Visual Analytics Framework for Multi-Faceted Understanding and Probing of Graph Neural Network Predictions