DynGEM: Deep Embedding Method for Dynamic Graphs

Abstract

Embedding large graphs in low dimensional spaces has recently attracted significant interest due to its wide applications such as graph visualization, link prediction and node classification. Existing methods focus on computing the embedding for static graphs. However, many graphs in practical applications are dynamic and evolve constantly over time. Naively applying existing embedding algorithms to each snapshot of dynamic graphs independently usually leads to unsatisfactory performance in terms of stability, flexibility and efficiency. In this work, we present an efficient algorithm DynGEM based on recent advances in deep autoencoders for graph embeddings, to address this problem. The major advantages of DynGEM include: (1) the embedding is stable over time, (2) it can handle growing dynamic graphs, and (3) it has better running time than using static embedding methods on each snapshot of a dynamic graph. We test DynGEM on a variety of tasks including graph visualization, graph reconstruction, link prediction and anomaly detection (on both synthetic and real datasets). Experimental results demonstrate the superior stability and scalability of our approach.

3rd International Workshop on Representation Learning for Graphs (ReLiG), IJCAI 2017

References in corpus (5)

• Graph Embedding Techniques, Applications, and Performance: A Survey
• Learning Convolutional Neural Networks for Graphs
• Diachronic Word Embeddings Reveal Statistical Laws of Semantic Change
• Net2Net: Accelerating Learning via Knowledge Transfer
• Deep Coevolutionary Network: Embedding User and Item Features for Recommendation

Cited by in corpus (74)

• Graph Embedding Techniques, Applications, and Performance: A Survey
• dyngraph2vec: Capturing Network Dynamics using Dynamic Graph Representation Learning
• Node Embedding over Temporal Graphs
• Graph Neural Networks: Taxonomy, Advances and Trends
• Temporal Graph Networks for Deep Learning on Dynamic Graphs
• EvoKG: Jointly Modeling Event Time and Network Structure for Reasoning over Temporal Knowledge Graphs
• TREND: TempoRal Event and Node Dynamics for Graph Representation Learning
• K-Core based Temporal Graph Convolutional Network for Dynamic Graphs
• Representation Learning for Dynamic Graphs: A Survey
• Representation Learning over Dynamic Graphs
• GloDyNE: Global Topology Preserving Dynamic Network Embedding
• Continuous-Time and Multi-Level Graph Representation Learning for Origin-Destination Demand Prediction
• CGC: Contrastive Graph Clustering for Community Detection and Tracking
• Recurrent Event Network: Autoregressive Structure Inference over Temporal Knowledge Graphs
• A Spatial-Temporal Graph Neural Network Framework for Automated Software Bug Triaging
• Dynamic Graph Convolutional Networks Using the Tensor M-Product
• Deep learning for dynamic graphs: models and benchmarks
• Inductive Representation Learning on Temporal Graphs
• A Survey on Dynamic Network Embedding
• Lifelong Learning of Graph Neural Networks for Open-World Node Classification
• PANE: scalable and effective attributed network embedding
• TCL: Transformer-based Dynamic Graph Modelling via Contrastive Learning
• Streaming Graph Neural Networks
• Subset Node Anomaly Tracking over Large Dynamic Graphs
• CLDG: Contrastive Learning on Dynamic Graphs
• Lifelong Learning on Evolving Graphs Under the Constraints of Imbalanced Classes and New Classes
• TG-GAN: Continuous-time Temporal Graph Generation with Deep Generative Models
• Learning Dynamic Embeddings from Temporal Interactions
• Learning Backward Compatible Embeddings
• DynamicGEM: A Library for Dynamic Graph Embedding Methods
• Understanding graph embedding methods and their applications
• Structural Temporal Graph Neural Networks for Anomaly Detection in Dynamic Graphs
• Embedding and trajectories of temporal networks
• Graph Convolutional Networks for traffic anomaly
• DyGCN: Dynamic Graph Embedding with Graph Convolutional Network
• On the Equivalence Between Temporal and Static Graph Representations for Observational Predictions
• Modeling Dynamic Heterogeneous Network for Link Prediction using Hierarchical Attention with Temporal RNN
• Temporal Attribute Prediction via Joint Modeling of Multi-Relational Structure Evolution
• Link prediction in dynamic networks using random dot product graphs
• DyANE: Dynamics-aware node embedding for temporal networks
• Revisiting Latent-Space Interpolation via a Quantitative Evaluation Framework
• Dynamic Network Embeddings for Network Evolution Analysis
• EvoNet: A Neural Network for Predicting the Evolution of Dynamic Graphs
• Laplacian-Based Dimensionality Reduction Including Spectral Clustering, Laplacian Eigenmap, Locality Preserving Projection, Graph Embedding, and Diffusion Map: Tutorial and Survey
• dynnode2vec: Scalable Dynamic Network Embedding
• DynWalks: Global Topology and Recent Changes Awareness Dynamic Network Embedding
• DyFormer: A Scalable Dynamic Graph Transformer with Provable Benefits on Generalization Ability
• Accelerating Dynamic Network Embedding with Billions of Parameter Updates to Milliseconds
• Graph Embedding Based Hybrid Social Recommendation System
• GRADE: Graph Dynamic Embedding
• ConTIG: Continuous Representation Learning on Temporal Interaction Graphs
• TSAM: Temporal Link Prediction in Directed Networks based on Self-Attention Mechanism
• Dynamic Graph Collaborative Filtering
• From Static to Dynamic Node Embeddings
• ForeSeer: Product Aspect Forecasting Using Temporal Graph Embedding
• Graph-Level Embedding for Time-Evolving Graphs
• Dynamic Sequential Graph Learning for Click-Through Rate Prediction
• Representing Social Networks as Dynamic Heterogeneous Graphs
• Temporal Graph Network Embedding with Causal Anonymous Walks Representations
• Temporal Network Embedding with Micro- and Macro-dynamics
• Robust Dynamic Network Embedding via Ensembles
• Multi-Aspect Temporal Network Embedding: A Mixture of Hawkes Process View
• Dynamic Graph Embedding via LSTM History Tracking
• Multi-View Dynamic Heterogeneous Information Network Embedding
• Embedding Dynamic Attributed Networks by Modeling the Evolution Processes
• FeatureNorm: L2 Feature Normalization for Dynamic Graph Embedding
• Online Dynamic Network Embedding
• Bitcoin Transaction Forecasting with Deep Network Representation Learning
• Learning Based Proximity Matrix Factorization for Node Embedding
• SPAN: Subgraph Prediction Attention Network for Dynamic Graphs
• Dynamic Embedding on Textual Networks via a Gaussian Process
• Spectral Transform Forms Scalable Transformer
• Dynamic Joint Variational Graph Autoencoders
• Temporal Graph Offset Reconstruction: Towards Temporally Robust Graph Representation Learning