Awesome GNN Research

🔥 Top AI&ML&DM Conferences or Journals and arXiv Prep.

CVPR, ICCV, ECCV, ACL, MM, etc

AAAI, IJCAI, SIGIR, ICDE, SIGMOD, etc

KDD, WWW, ICLR, ICML, NeurIPS, TKDE, etc

⭐ Research Keywords

Scalable Graph Neural Networks, Federated Graph Learning, Recommender system Based on GNN

Wizardship

• Scalable Graph Neural Networks
  • Graph Embedding
  • Linear-model based Graph Neural Networks
  • Sampling based Graph Neural Networks
  • Model Compression and Quantification
  • Efficient Architecture and Paradigm
  • Graph Data Augmentation
  • Imbalance Graph Neural Networks
• Federated Graph Learning
  • Personalized and Heterogeneous Federated Learning in CV or NLP
  • Theoretical Analysis of Federated Learning in CV or NLP
  • CV or NLP Model Compression and Quantification in Federated Learning
  • Transfer Federated Graph Learning and Graph Structure Federated Learning
  • Intra-Graph Horizontal Federated Learning
  • Inter-Graph Horizontal Federated Learning
  • Vertical Federal Learning
• Privacy Graph Neural Networks
• Survey and Framework Toolkits

Scalable Graph Neural Networks

Graph Embedding

• KDD'14 DeepWalk: Online Learning of Social Representations [Paper] [Code] [Link]
• WWW'15 LINE: Large-scale Information Network Embedding [Paper] [Code] [Link]
• KDD'16 node2vec: Scalable Feature Learning for Networks [Paper] [Code] [Link]
• NeurIPS'13 Distributed Representations of Words and Phrases and their Compositionality [Paper] [Code] [Link]
• KDD'16 Structural Deep Network Embedding [Paper] [Code] [Link]

Linear-model based Graph Neural Networks

• ICML'19 Simplifying Graph Convolutional Networks [Paper] [Code] [Link]
• ICLR'19 Predict Then Propagate: Graph Neural Networks Meet Personalized PageRank [Paper] [Code] [Link]
• arXiv'20 Scalable Graph Neural Networks for Heterogeneous Graphs [Paper] [Code] [Link]
• arXiv'20 Unifying Graph Convolutional Neural Networks and Label Propagation [Paper] [Code] [Link]
• NeurIPS‘21 Node Dependent Local Smoothing for Scalable Graph Learning [Paper] [Code] [Link]
• arXiv'21 Scalable and Adaptive Graph Neural Networks with Self-Label-Enhanced Training [Paper] [Code] [Link]
• ICLR'21 Combining Label Propagation and Simple Models Out-performs Graph Neural Networks [Paper] [Code] [Link]
• ICLR'22 Node Feature Extraction by Self-Supervised Multi-scale Neighborhood Prediction [Paper] [Code] [Link]
• arXiv'22 SCR: Training Graph Neural Networks with Consistency Regularization [Paper] [Code] [Link]
• KDD'22 Graph Attention MLP with Reliable Label Utilization [Paper] [Code] [Link]

Sampling based Graph Neural Networks

• NeurIPS'17 Inductive Representation Learning on Large Graphs [Paper] [Code] [Link]
• ICLR'18 FASTGCN: Fast Learning With Graph Convolutional Networks Via Importance Sampling [Paper] [Code] [Link]

Model Compression and Quantification

• NeurIPS'14 Distilling the Knowledge in a Neural Network [Paper] [Code] [Link]
• ICLR'15 FitNets: Hints for Thin Deep Nets [Paper] [No Code] [Link]
• ICLR'17 Paying More Attention to Attention: Improving The Performance of Convolutional Neural Networks via Attention Transfer [Paper] [Code] [Link]
• ICCV'19 Similarity-Preserving Knowledge Distillation [Paper] [No Code] [Link]
• ICCV'19 Be Your Own Teacher: Improve the Performance of Convolutional Neural Networks via Self Distillation [Paper] [Code] [Link]
• CVPR'21 Distilling Knowledge via Knowledge Review [Paper] [Code] [Link]
• CVPR'21 Distill on the Go: Online knowledge distillation in self-supervised learning [Paper] [Code] [Link]
• CVPR'22 Decoupled Knowledge Distillation [Paper] [Code] [Link]

---

• CVPR'20 Distilling Knowledge from Graph Convolutional Networks [Paper] [Code] [Link]
• SIGMOD'20 Reliable Data Distillation on Graph Convolutional Network [Paper] [No Code] [Link]
• KDD'20 TinyGNN: Learning Efficient Graph Neural Networks [Paper] [No Code] [Link]
• arXiv’21 Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages [Paper] [Code] [Link]
• MICCAI'21 GKD: Semi-supervised Graph Knowledge Distillation for Graph-Independent Inference [Paper] [Code] [Link]
• CVPR'21 Bi-GCN: Binary Graph Convolutional Network [Paper] [Code] [Link]
• IJCAI'21 Graph-Free Knowledge Distillation for Graph Neural Networks [Paper] [Code] [Link]
• IJCAI'21 On Self-Distilling Graph Neural Network [Paper] [No Code] [Link]
• ICLR'21 On Graph Neural Networks versus Graph-Augmented MLPs [Paperr] [Code] [Link]
• WWW'21 Extract the Knowledge of Graph Neural Networks and Go Beyond it: An Effective Knowledge Distillation Framework [Paper] [Code] [Link]
• KDD'21 ROD: Reception-aware Online Distillation for Sparse Graphs [Paper] [Code] [Link]
• arXiv'22 On Representation Knowledge Distillation for Graph Neural Networks [Paper] [No Code] [Link]
• AAAI'22 Workshop Scalable Consistency Training for Graph Neural Networks via Self-Ensemble Self-Distillation [Paper] [No Code] [Link]
• ICLR'22 Graph-less Neural Networks: Teaching Old MLPs New Tricks via Distillation [Paper] [Code] [Link]
• KDD'22 Compressing Deep Graph Neural Networks via Adversarial Knowledge Distillation [Paper] [Code] [Link]
• ICLR'22 Cold Brew Distilling Graph Node Representations with Incomplete or Missing Neighborhoods [Paper] [Code] [Link]
• WSDM'22 Deep Graph-level Anomaly Detection by Glocal Knowledge Distillation [Paper] [Code] [Link]

Efficient Architecture and Paradigm

• ICLR'19 How Powerful are Graph Neural Networks? [Paper] [Code] [Link]
• NeurIPS'20 Graph Random Neural Networks for Semi-Supervised Learning on Graphs [Paper] [Code] [Link]
• arXiv'21 Graph Learning with 1D Convolutions on Random Walks [Paper] [Code] [Link]
• WSDM'21 Node Similarity Preserving Graph Convolutional Networks [Paper] [Code] [Link]
• ICLR'22 A New Perspective on "How Graph Neural Networks Go Beyond Weisfeiler-Lehman?" [Paper] [No Code] [Link]
• KDD'22 Model Degradation Hinders Deep Graph Neural Networks [Paper] [Code] [Link]
• KDD'22 Feature Overcorrelation in Deep Graph Neural Networks: A New Perspective [Paper] [Code] [Link]

Graph Data Augmentation

• KDD'20 NodeAug: Semi-Supervised Node Classification with Data Augmentation [Paper] [No Code] [Link]
• arXiv'21 Local Augmentation for Graph Neural Networks [Paper] [No Code] [Link]
• AAAI'21 Data Augmentation for Graph Neural Networks [Paper] [Code] [Link]
• WWW'21 Graph Contrastive Learning with Adaptive Augmentation [Paper] [Code] [Link]
• AAAI'22 Regularizing Graph Neural Networks via Consistency-Diversity Graph Augmentations [Paper] [No Code] [Link]
• AAAI'21 GraphMix: Improved Training of GNNs for Semi-Supervised Learning [Paper] [Code] [Link]
• CVPR'22 Robust Optimization as Data Augmentation for Large-scale Graphs [Paper] [Code] [Link]
• AAAI'22 SAIL: Self-Augmented Graph Contrastive Learning [Paper] [No Code] [Link]

Imbalance Graph Neural Networks

• arXiv'20 Non-Local Graph Neural Networks [Paper] [No Code] [Link]
• arXiv'20 Non-IID Graph Neural Networks [Paper] [No Code] [Link]
• WSDM'21 GraphSMOTE: Imbalanced Node Classification on Graphs with Graph Neural Networks [Paper] [Code] [Link]
• IJCAI'21 Multi-Class Imbalanced Graph Convolutional Network Learning [Paper] [Code] [Link]
• NeurIPS’21 Shift-Robust GNNs: Overcoming the Limitations of Localized Graph Training Data [Paper] [Code] [Link]

Federated Graph Learning

• Big Data'19 SGNN: A Graph Neural Network Based Federated Learning Approach by Hiding Structure [Paper] [No Code] [Link]

• arXiv'20 Federated Dynamic GNN with Secure Aggregation [Paper] [No Code] [Link]

• arXiv'21 GIST: Distributed Training for Large-Scale Graph Convolutional Networks [Paper] [Code] [Link]

• TSIPN'21 Distributed Training of Graph Convolutional Networks [Paper] [No Code] [Link]

• ICLR'22 Learn Locally, Correct Globally: A Distributed Algorithm for Training Graph Neural Networks [Paper] [Code] [Link]

Personalized and Heterogeneous Federated Learning in CV or NLP

• NeurIPS'18 Workshop Communication-Efficient On-Device Machine Learning: Federated Distillation and Augmentation under Non-IID Private Data [Paper] [No Code] [Link]
• NeurIPS'19 Think Locally, Act Globally: Federated Learning with Local and Global Representations [Paper] [Code] [Link]
• arXiv'20 Adaptive Personalized Federated Learning [Paper] [No Code] [Link]
• AAAI'21 Addressing Class Imbalance in Federated Learning [Paper] [Code] [Link]

Theoretical Analysis of Federated Learning in CV or NLP

• JMLR'17 Communication-Efficient Learning of Deep Networks from Decentralized Data [Paper] [Code] [Link]
• arXiv'19 Detailed comparison of communication efficiency of split learning and federated learning [Paper] [Link]
• ICLR'20 On the Convergence of FedAvg on Non-IID Data [Paper] [Code] [Link]

CV or NLP Model Compression and Quantification in Federated Learning

• NeurIPS'19 Workshop FedMD: Heterogenous Federated Learning via Model Distillation [Paper] [Code] [Link]
• NeurIPS'20 Group Knowledge Transfer: Federated Learning of Large CNNs at the Edge [Paper] [Code] [Link]
• arXiv'22 CDKT-FL: Cross-Device Knowledge Transfer using Proxy Dataset in Federated Learning [Paper] [Code] [Link]

Transfer Federated Graph Learning and Graph Structure Federated Learning

• arXiv'19 Peer-to-Peer Federated Learning on Graphs [Paper] [No Code] [Link]
• ICML'21 SpreadGNN: Serverless Multi task Federated Learning for Graph Neural Networks [Paper] [Code] [Link]
• PPNA'21 ASFGNN: Automated Separated-Federated Graph Neural Network [Paper] [No Code] [Link]
• IJCAI'21 Decentralized Federated Graph Neural Networks [Paper] [No Code] [Link]
• CVPR'21 Cluster-driven Graph Federated Learning over Multiple Domains [Paper] [No Code] [Link]
• ICML'21 Personalized Federated Learning using Hypernetworks [Paper] [Code] [Link]

Intra-Graph Horizontal Federated Learning

• arXiv'20 GraphFL: A Federated Learning Framework for Semi-Supervised Node Classification on Graphs [Paper] [No Code] [Link]
• NeurIPS'21 Subgraph Federated Learning with Missing Neighbor Generation [Paper] [Code] [Link]
• ICML'21 FedGNN: Federated Graph Neural Network for Privacy-Preserving Recommendation [Paper] [No Code] [Link]
• arXiv'21 FedGL: Federated Graph Learning Framework with Global Self-Supervision [Paper] [No Code] [Link]
• KDD'21 Cross-Node Federated Graph Neural Network for Spatio-Temporal Data Modeling [Paper] [Code] [Link]
• CIKM'21 Federated Knowledge Graphs Embedding [Paper] [Code] [Link]

Inter-Graph Horizontal Federated Learning

• NeurIPS'21 Federated Graph Classification over Non-IID Graphs [Paper] [Code] [Link]

Vertical Federal Learning

• arXiv'21 A Vertical Federated Learning Framework for Graph Convolutional Network [Paper] [No Code] [Link]
• arXiv'21 Vertically Federated Graph Neural Network for Privacy-Preserving Node Classification [Paper] [No Code] [Link]

Privacy Graph Neural Networks

• SIGSAC'16 Deep Learning with Differential Privacy [Paper] [Code] [Link]
• ICLR'17 Semi-supervised Knowledge Transfer for Deep Learning from Private Training Data [Paper] [Code] [Link]
• ICLR'18 Scalable Private Learning With PATE [Paper] [Code] [Link]
• arXiv'20 When Differential Privacy Meets Graph Neural Networks [Paper] [Code] [Link]
• arXiv'21 Releasing Graph Neural Networks with Differential Privacy [Paper] [No Code] [Link]
• arXiv'21 A Graph Federated Architecture with Privacy Preserving Learning [Paper] [No Code] [Link]
• IJCAI'21 Secure Deep Graph Generation with Link Differential Privacy [Paper] [Code] [Link]
• CCS'21 Locally Private Graph Neural Networks [Paper] [Code] [Link]

Survey and Framework Toolkits

• Graph library -- PyG、GarphGallery [Link]

• Graph library -- DIG、AutoGL、CogDL [Link]

• PyTorch Geometric（一）：数据加载 [Link]

• PyTorch Geometric（二）：模型搭建 [Link]

• 基于 GNN 的隐私计算（联邦学习）Review（二）[Link]

• 基于 GNN 的隐私计算（联邦学习）Review（三）[Link]

• introduction [Link]

• Local Differential Privacy: a tutorial [Paper] [Link]

• 本地化差分隐私研究综述 [Paper] [Link]

• 差分隐私 -- Laplace mechanism、Gaussian mechanism、Composition theorem [Link]

• 矩母函数 GMF 及矩的概念 -- 期望、方差、归一化矩、偏态、峰度 [Link] [Reference]

• Moments Accountant 的理解 [Link] [Reference]

• 基于 GNN 的隐私计算（差分隐私）Review（一）[Link]

• Federated Machine Learning: Concept and Applications [Paper] [Link]

• arXiv'21 Graph4Rec: A Universal Toolkit with Graph Neural Networks for Recommender Systems [Paper] [Code] [Link]

• arXiv'21 Federated Graph Learning - A Position Paper [Paper] [Link]

• arXiv'21 Federated Learning on Non-IID Data Silos: An Experimental Study [Paper] [Code] [Link]

• ICLR'21 FedGraphNN: A Federated Learning Benchmark System for Graph Neural Networks [Paper] [Code] [Link]

• arXiv‘22 Data Augmentation for Deep Graph Learning: A Survey [Paper] [No Code] [Link]

• arXiv'22 A Survey on Graph Structure Learning: Progress and Opportunities [Paper] [No Code] [Link]