Subgraph Neural Networks

2020/06/18 by Emily Alsentzer, Alsentzer, Emily, Samuel G. Finlayson +6 · 9 citations
Biochemistry, Genetics and Molecular Biology · Computer Science · Mathematics · Physics and Astronomy · #Advanced Graph Neural Networks #Bioinformatics and Genomic Networks #Complex Network Analysis Techniques #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Social and Information Networks (cs.SI) #cs.LG #cs.SI #stat.ML

paper · pdf · doi:10.48550/arxiv.2006.10538

E.A. and S.G.F. contributed equally

openalex publication_date 2020/06/18 · arxiv created 2020/11/06 · arxiv updated 2020/11/10 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Deep learning methods for graphs achieve remarkable performance on many node-level and graph-level prediction tasks. However, despite the proliferation of the methods and their success, prevailing Graph Neural Networks (GNNs) neglect subgraphs, rendering subgraph prediction tasks challenging to tackle in many impactful applications. Further, subgraph prediction tasks present several unique challenges: subgraphs can have non-trivial internal topology, but also carry a notion of position and external connectivity information relative to the underlying graph in which they exist. Here, we introduce SubGNN, a subgraph neural network to learn disentangled subgraph representations. We propose a novel subgraph routing mechanism that propagates neural messages between the subgraph's components and randomly sampled anchor patches from the underlying graph, yielding highly accurate subgraph representations. SubGNN specifies three channels, each designed to capture a distinct aspect of subgraph topology, and we provide empirical evidence that the channels encode their intended properties. We design a series of new synthetic and real-world subgraph datasets. Empirical results for subgraph classification on eight datasets show that SubGNN achieves considerable performance gains, outperforming strong baseline methods, including node-level and graph-level GNNs, by 19.8% over the strongest baseline. SubGNN performs exceptionally well on challenging biomedical datasets where subgraphs have complex topology and even comprise multiple disconnected components.

Citations

• Gene Ontology: tool for the unification of biology
• Network structure and minimum degree
• Learning Convolutional Neural Networks for Graphs
• Neural Message Passing for Quantum Chemistry
• Inductive Representation Learning on Large Graphs
• Learning Graph-Level Representation for Drug Discovery
• Graph Neural Networks: A Review of Methods and Applications
• Fast Graph Representation Learning with PyTorch Geometric
• MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing
• Position-aware Graph Neural Networks
• Detecting Functionality-Specific Vulnerabilities via Retrieving Individual Functionality-Equivalent APIs in Open-Source Repositories
• vGraph: A Generative Model for Joint Community Detection and Node Representation Learning
• Strategies for Pre-training Graph Neural Networks
• A Comprehensive Survey on Graph Neural Networks

Cited by

• A Multimodal Approach to Alzheimer's Diagnosis: Geometric Insights from Cube Copying and Cognitive Assessments
• GRAVER: Generative Graph Vocabularies for Robust Graph Foundation Models Fine-tuning
• BioBlobs: Unsupervised Discovery of Functional Substructures for Protein Function Prediction
• M4GN: Mesh-based Multi-segment Hierarchical Graph Network for Dynamic Simulations
• SlepNet: Spectral Subgraph Representation Learning for Neural Dynamics
• Soft causal learning for generalized molecule property prediction: An environment perspective
• On Universality Classes of Equivariant Networks
• Clustering and analysis of user behaviour in blockchain: A case study of Planet IX
• BEACON: A Benchmark for Efficient and Accurate Counting of Subgraphs

Related

• Using Subgraph GNNs for Node Classification:an Overlooked Potential Approach
• Deep Generative Models for Subgraph Prediction
• SUGAR: Subgraph Neural Network with Reinforcement Pooling and Self-Supervised Mutual Information Mechanism
• Translating Subgraphs to Nodes Makes Simple GNNs Strong and Efficient for Subgraph Representation Learning
• Self-supervised Subgraph Neural Network With Deep Reinforcement Walk Exploration