Dynamic Graph/Network Embedding

离散时间

• DySAT: Deep Neural Representation Learning on Dynamic Graphs via Self-Attention Networks

  • Paper
  • Code
  • WSDM' 20
  • GNN + self-attention

• DyGGNN: Learning to Represent the Evolution of Dynamic Graphs with Recurrent Models

  • Paper
  • WWW' 19
  • GGNN + LSTM

• CTDNE: Continuous-Time Dynamic Network Embeddings

  • Paper
  • Code
  • WWW' 19
  • 基于随机游走

• EvolveGCN: Evolving Graph Convolutional Networks for Dynamic Graphs

  • Paper
  • Code
  • AAAI' 20

• DynGEM: A Library for Dynamic Graph Embedding Methods

  • Paper
  • Code
  • IJCAI' 17 workshop

• DGNN: Streaming Graph Neural Networks

  • Paper
  • SIGIR' 20
  • 基于 LSTM 对更新组件进行建模，由两个主要部分组成：更新组件和传播组件，不考虑边和节点的删除

• DynGraph2Seq: Dynamic-Graph-to-Sequence Interpretable Learning for Health Stage Prediction in Online Health Forums

  • Paper
  • ICDM’ 2019

• TMF: Temporally Factorized Network Modeling for Evolutionary Network Analysis

  • Paper
  • WSDM' 17
  • 矩阵分解，将网络的边缘结构纯粹表示为时间的函数，并预测网络随时间的演变（节点不能增删）

• DynamicTriad: Dynamic Network Embedding by Modeling Triadic Closure Process

  • Paper
  • Code
  • AAAI' 18
  • 无向图

• NAAM: Learning from Dynamic User Interaction Graphs to Forecast Diverse Social Behavior

  • Paper
  • CIKM' 19
  • GCN+RNN/LSTM

• Know-Evolve

  • Paper
  • ICML' 17

• DyGCN

  • Paper

  • code

  • Woodstock' 18

Survey

• A Survey on Embedding Dynamic Graphs

  • Paper
  • CSUR

• Representation Learning for Dynamic Graphs: A Survey

  • Paper