Explainability Techniques for Graph Convolutional Networks

Abstract

Graph Networks are used to make decisions in potentially complex scenarios but it is usually not obvious how or why they made them. In this work, we study the explainability of Graph Network decisions using two main classes of techniques, gradient-based and decomposition-based, on a toy dataset and a chemistry task. Our study sets the ground for future development as well as application to real-world problems.

Accepted at the ICML 2019 Workshop "Learning and Reasoning with Graph-Structured Representations" (poster + spotlight talk)

References in corpus (13)

• Neural Message Passing for Quantum Chemistry
• Axiomatic Attribution for Deep Networks
• Learning Important Features Through Propagating Activation Differences
• Molecular Graph Convolutions: Moving Beyond Fingerprints
• Explaining NonLinear Classification Decisions with Deep Taylor Decomposition
• Variational Graph Auto-Encoders
• SmoothGrad: removing noise by adding noise
• How to Explain Individual Classification Decisions
• Interaction Networks for Learning about Objects, Relations and Physics
• Learning how to explain neural networks: PatternNet and PatternAttribution
• A Compositional Object-Based Approach to Learning Physical Dynamics
• Multifaceted Feature Visualization: Uncovering the Different Types of Features Learned By Each Neuron in Deep Neural Networks
• Interpreting Neural Network Judgments via Minimal, Stable, and Symbolic Corrections

Cited by in corpus (17)

• Graph Neural Networks: A Review of Methods and Applications
• Graph Neural Network for Traffic Forecasting: A Survey
• Graph-Based Deep Learning for Medical Diagnosis and Analysis: Past, Present and Future
• Graph Neural Networks: Taxonomy, Advances and Trends
• Interpreting Graph Neural Networks for NLP With Differentiable Edge Masking
• Bridging the Gap between Spatial and Spectral Domains: A Survey on Graph Neural Networks
• Towards Human-like Perception: Learning Structural Causal Model in Heterogeneous Graph
• Guiding Cascading Failure Search with Interpretable Graph Convolutional Network
• GCExplainer: Human-in-the-Loop Concept-based Explanations for Graph Neural Networks
• Sampling methods for efficient training of graph convolutional networks: A survey
• Deep Graph Similarity Learning: A Survey
• Contrastive Graph Neural Network Explanation
• Deep Ensembles from a Bayesian Perspective
• Preserve, Promote, or Attack? GNN Explanation via Topology Perturbation
• Explain Graph Neural Networks to Understand Weighted Graph Features in Node Classification
• Financial Crime & Fraud Detection Using Graph Computing: Application Considerations & Outlook
• LW-GCN: A Lightweight FPGA-based Graph Convolutional Network Accelerator