Deep Learning on Graphs: A Survey

Abstract

Deep learning has been shown to be successful in a number of domains, ranging from acoustics, images, to natural language processing. However, applying deep learning to the ubiquitous graph data is non-trivial because of the unique characteristics of graphs. Recently, substantial research efforts have been devoted to applying deep learning methods to graphs, resulting in beneficial advances in graph analysis techniques. In this survey, we comprehensively review the different types of deep learning methods on graphs. We divide the existing methods into five categories based on their model architectures and training strategies: graph recurrent neural networks, graph convolutional networks, graph autoencoders, graph reinforcement learning, and graph adversarial methods. We then provide a comprehensive overview of these methods in a systematic manner mainly by following their development history. We also analyze the differences and compositions of different methods. Finally, we briefly outline the applications in which they have been used and discuss potential future research directions.

arXiv commentAccepted by Transactions on Knowledge and Data Engineering. 24 pages, 11 figures

References in corpus (41)

• Emergence of scaling in random networks
• Auto-Encoding Variational Bayes
• The Emerging Field of Signal Processing on Graphs: Extending High-Dimensional Data Analysis to Networks and Other Irregular Domains
• Inductive Representation Learning on Large Graphs
• Neural Message Passing for Quantum Chemistry
• Geometric deep learning: going beyond Euclidean data
• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• Molecular Graph Convolutions: Moving Beyond Fingerprints
• Modeling polypharmacy side effects with graph convolutional networks
• Graph Neural Networks: A Review of Methods and Applications
• Deep Convolutional Networks on Graph-Structured Data
• Fast Graph Representation Learning with PyTorch Geometric
• Simplifying Graph Convolutional Networks
• Graph Convolutional Matrix Completion
• Adversarial Attacks on Neural Networks for Graph Data
• Representation Learning on Graphs with Jumping Knowledge Networks
• Variational Graph Auto-Encoders
• Pitfalls of Graph Neural Network Evaluation
• Interaction Networks for Learning about Objects, Relations and Physics
• Discriminative Embeddings of Latent Variable Models for Structured Data
• GraphRNN: Generating Realistic Graphs with Deep Auto-regressive Models
• MolGAN: An implicit generative model for small molecular graphs
• Dynamic Graph Convolutional Networks
• Learning Deep Generative Models of Graphs
• Few-Shot Learning with Graph Neural Networks
• Go for a Walk and Arrive at the Answer: Reasoning Over Paths in Knowledge Bases using Reinforcement Learning
• Adversarial Attacks on Graph Neural Networks via Meta Learning
• Combinatorial Optimization with Graph Convolutional Networks and Guided Tree Search
• Adversarial Attack and Defense on Graph Data: A Survey
• Graph Isomorphism in Quasipolynomial Time
• Revisiting Graph Neural Networks: All We Have is Low-Pass Filters
• Adaptive Sampling Towards Fast Graph Representation Learning
• NetGAN: Generating Graphs via Random Walks
• Deep Neural Networks with Random Gaussian Weights: A Universal Classification Strategy?
• Learning Disentangled Representations for Recommendation
• Graph Wavelet Neural Network
• Structural Deep Embedding for Hyper-Networks
• Signed Graph Convolutional Network
• Towards Gene Expression Convolutions using Gene Interaction Graphs
• Streaming Graph Neural Networks
• A Note on Spectral Clustering and SVD of Graph Data

Cited by in corpus (48)

• Convergence of Edge Computing and Deep Learning: A Comprehensive Survey
• GNNExplainer: Generating Explanations for Graph Neural Networks
• Foundations and modelling of dynamic networks using Dynamic Graph Neural Networks: A survey
• A Gentle Introduction to Deep Learning for Graphs
• SIGN: Scalable Inception Graph Neural Networks
• Graph Neural Networks: Taxonomy, Advances and Trends
• Examining COVID-19 Forecasting using Spatio-Temporal Graph Neural Networks
• Graph Convolutional Networks for Graphs Containing Missing Features
• Hierarchical Graph Pooling with Structure Learning
• Graph Neural Networks Meet Neural-Symbolic Computing: A Survey and Perspective
• Adversarial Attacks Against Deep Generative Models on Data: A Survey
• Hypergraph Convolution and Hypergraph Attention
• Bridging the Gap between Spatial and Spectral Domains: A Survey on Graph Neural Networks
• Edge-Featured Graph Attention Network
• Graph Adversarial Training: Dynamically Regularizing Based on Graph Structure
• Finding Patient Zero: Learning Contagion Source with Graph Neural Networks
• Discrete Graph Structure Learning for Forecasting Multiple Time Series
• Label Efficient Semi-Supervised Learning via Graph Filtering
• Deep Learning for Insider Threat Detection: Review, Challenges and Opportunities
• Learning to Drop: Robust Graph Neural Network via Topological Denoising
• Bilinear Graph Neural Network with Neighbor Interactions
• Multitask Learning on Graph Neural Networks: Learning Multiple Graph Centrality Measures with a Unified Network
• GraphConfRec: A Graph Neural Network-Based Conference Recommender System
• Graph Convolutional Networks with EigenPooling
• Rubik: A Hierarchical Architecture for Efficient Graph Learning
• Relational Graph Representation Learning for Open-Domain Question Answering
• Adaptive-Step Graph Meta-Learner for Few-Shot Graph Classification
• FrequentNet: A Novel Interpretable Deep Learning Model for Image Classification
• Independence Promoted Graph Disentangled Networks
• Next Waves in Veridical Network Embedding
• Explain Graph Neural Networks to Understand Weighted Graph Features in Node Classification
• GraphTER: Unsupervised Learning of Graph Transformation Equivariant Representations via Auto-Encoding Node-wise Transformations
• Dynamic Graph Representation for Partially Occluded Biometrics
• Modeling Edge Features with Deep Bayesian Graph Networks
• Deep Generative Modeling in Network Science with Applications to Public Policy Research
• A Graph Convolutional Network Composition Framework for Semi-supervised Classification
• HiCoRe: Visual Hierarchical Context-Reasoning
• Multi-Level Attention Pooling for Graph Neural Networks: Unifying Graph Representations with Multiple Localities
• Lorentzian Graph Convolutional Networks
• Deep Neural Network for 3D Surface Segmentation based on Contour Tree Hierarchy
• Fast Haar Transforms for Graph Neural Networks
• Community-preserving Graph Convolutions for Structural and Functional Joint Embedding of Brain Networks
• Mapping the Internet: Modelling Entity Interactions in Complex Heterogeneous Networks
• Decoupled Variational Embedding for Signed Directed Networks
• A Method to Predict Semantic Relations on Artificial Intelligence Papers
• Graph-Preserving Grid Layout: A Simple Graph Drawing Method for Graph Classification using CNNs
• Supervised learning on heterogeneous, attributed entities interacting over time
• Graph Minors Meet Machine Learning: the Power of Obstructions