Graph Data Augmentation (GraphDA) for Deep Graph Learning

The repository contains links primarily to conference and journal publications about graph data augmentation for deep graph learning. If you find this repository useful, please kindly cite the following paper: Data Augmentation for Deep Graph Learning: A Survey

@article{ding2022data,
  title={Data Augmentation for Deep Graph Learning: A Survey},
  author={Ding, Kaize and Xu, Zhe and Tong, Hanghang and Liu, Huan},
  journal={arXiv preprint arXiv:2202.08235},
  year={2022}
}

We will keep updating the paper list and you are highly encouraged to contribute to this repo!

Roadmaps

• GraphDA for Optimal Graph Learning
  • Optimal Structure Learning
  • Optimal Feature Learning
• GraphDA for Low-Resource Graph Learning
  • Graph Self-Supervised Learning
  • Graph Self/Co-Training
  • Graph Interpolation
  • Graph Consistency Training
• Other Directions
  • GraphDA for Robust Graph Learning
  • GraphDA for Graph Imbalanced Learning
  • GraphDA for Learning on Heterophilic Graphs

GraphDA for Optimal Graph Learning

Optimal Structure Learning

Name	Paper
Gasoline	[TheWebConf 2022] Graph Sanitation with Application to Node Classification [Code]
GEN	[TheWebConf 2021] Graph Structure Estimation Neural Networks [Code]
PTDNet	[WSDM 2021] Learning to drop: Robust graph neural network via topological denoising [Code]
GAUG	[AAAI 2021] Data Augmentation for Graph Neural Networks [Code]
HGSL	[AAAI 2021] Heterogeneous Graph Structure Learning for Graph Neural Networks [Code]
ESML	[ICDE 2021] Edge Sparsification for Graphs via Meta-Learning
GIB-N	[NeurIPS 2020] Graph Information Bottleneck [Code]
IDGL	[NeurIPS 2020] Iterative deep graph learning for graph neural networks: Better and robust node embeddings [Code]
NeuralSparse	[ICML 2020] Robust graph representation learning via neural sparsification
Pro-GNN	[KDD 2020] Graph structure learning for robust graph neural networks [Code]
GIB-G	[ICLR 2020] Graph Information Bottleneck for Subgraph Recognition
AdaEdge	[AAAI 2020] Measuring and relieving the over-smoothing problem for graph neural networks from the topological view
DenSE	[ICDM 2020] Learning Node Representations from Noisy Graph Structures
Grale	[KDD 2020] Grale: Designing Networks for Graph Learning
TO-GNN	[IJCAI 2019] Topology Optimization based Graph Convolutional Network
LDS	[ICML 2019] Learning discrete structures for graph neural networks [Code]
BGCN	[AAAI 2019] Bayesian graph convolutional neural networks for semi-supervised classification [Code]
PG-LEARN	[CIKM 2018] A Quest for Structure: Jointly Learning the Graph Structure and Semi-Supervised Classification [Code]
DGM	[arXiv] Differentiable Graph Module (DGM) for Graph Convolutional Networks

Optimal Feature Learning

Name	Paper
AirGNN	[NeurIPS 2021] Graph Neural Networks with Adaptive Residual [Code]
FP	[arXiv 2021] On the Unreasonable Effectiveness of Feature propagation in Learning on Graphs with Missing Node Features
GCNMF	[FGCS 2021] Graph convolutional networks for graphs containing missing features

GraphDA for Low-Resource Graph Learning

Graph Self-Supervised Learning

Name	Paper
S^3-CL	[arXiv 2022] Structural and Semantic Contrastive Learning for Self-supervised Node Representation Learning
BGRL	[ICLR 2022] Large-Scale Representation Learning on Graphs via Bootstrapping [Code]
ARIEL	[TheWebConf 2022] Adversarial Graph Contrastive Learning with Information Regularization
AD-GCL	[NeurIPS 2021] Adversarial Graph Augmentation to Improve Graph Contrastive Learning [Code]
GCA	[TheWebConf 2021] Graph Contrastive Learning with Adaptive Augmentation [Code]
MERIT	[IJCAI 2021] Multi-Scale Contrastive Siamese Networks for Self-Supervised Graph Representation Learning [Code]
CSSL	[AAAI 2021] Contrastive self-supervised learning for graph classification
GraphCL	[NeurIPS 2020] Graph contrastive learning with augmentations [Code]
GPT-GNN	[KDD 2020] Gpt-gnn: Generative pre-training of graph neural networks [Code]
GCC	[KDD 2020] GCC: Graph contrastive coding for graph neural network pre-training [Code]
MTL	[ICML 2020] When does self-supervision help graph convolutional networks?] [Code]
GraphBert	[arXiv 2020] Graph-bert: Only attention is needed for learning graph representations
Pre-train	[arXiv 2019] Pre-training graph neural networks for generic structural feature extraction

Graph Self/Co-Training

Name	Paper
Meta-PN	[AAAI 2022] Meta Propagation Networks for Graph Few-shot Semi-supervised Learning [Code]
NRGNN	[KDD 2021] NRGNN: Learning a Label Noise-Resistant Graph Neural Network on Sparsely and Noisily Labeled Graphs [Code]
PTA	[TheWebConf 2021] On the equivalence of decoupled graph convolution network and label propagation [Code]
CGCN	[AAAI 2020] Collaborative graph convolutional networks: Unsupervised learning meets semi-supervised learning
M3S	[AAAI 2020] Multi-stage self-supervised learning for graph convolutional networks on graphs with few labeled nodes [Code]
Co-GCN	[AAAI 2020] Co-GCN for Multi-View Semi-Supervised Learning
ST-GCNs	[AAAI 2018] Deeper insights into graph convolutional networks for semi-supervised learning [Code]

Graph Interpolation

Name	Paper
Graph Transplant	[AAAI 2022] Graph Transplant: Node Saliency-Guided Graph Mixup with Local Structure Preservation
G-Mixup	[TheWebConf 2021] Mixup for Node and Graph Classification [Code]
GraphMix	[AAAI 2021] GraphMix: Improved Training of GNNs for Semi-Supervised Learning [Code]
GraphMixup	[arXiv 2021] GraphMixup: Improving Class-Imbalanced Node Classification on Graphs by Self-supervised Context Prediction
ifMixup	[arXiv 2021] Intrusion-Free Graph Mixup

Graph Consistency Training

Name	Paper
MH-Aug	[NeurIPS 2021] Metropolis-Hastings Data Augmentation for Graph Neural Networks
GRAND	[NeurIPS 2020] Graph Random Neural Network for Semi-Supervised Learning on Graphs [Code]
NodeAug	[KDD 2020] NodeAug: Semi-Supervised Node Classification with Data Augmentation

Other Directions

GraphDA for Robust Graph Learning

Name	Paper
VGCN	[NeurIPS 2020] Variational Inference for Graph Convolutional Networks in the Absence of Graph Data and Adversarial Settings [Code]
GLNN	[ICME 2020] Exploring structure-adaptive graph learning for robust semi-supervised classification
GNN-Guard	[NeurIPS 2020] Gnnguard: Defending graph neural networks against adversarial attacks [Code]
FLAG	[arXiv] Flag: Adversarial data augmentation for graph neural networks
BVAT	[arXiv] Batch virtual adversarial training for graph convolutional networks
GraphAT	[TKDE 2019] Graph adversarial training: Dynamically regularizing based on graph structure [Code]
GCNVAT	[PRCV 2019] Virtual adversarial training on graph convolutional networks in node classification

More works on adversarial attack and defense on graphs can be found in this survey.

GraphDA for Graph Imbalanced Learning

Name	Paper
GraphSMOTE	[WSDM 2021] GraphSMOTE: Imbalanced Node Classification on Graphs with Graph Neural Networks [Code]
ImGAGN	[KDD 2021] ImGAGN: Imbalanced Network Embedding via Generative Adversarial Graph Networks [Code]
GraphMixup	[arXiv 2021] GraphMixup: Improving Class-Imbalanced Node Classification on Graphs by Self-supervised Context Prediction
DR-GCN	[IJCAI 2020] Multi-class imbalanced graph convolutional network learning

GraphDA for Learning on Heterophilic Graphs

Name	Paper
WRGAT	[KDD 2021] Breaking the Limit of Graph Neural Networks by Improving the Assortativity of Graphs with Local Mixing Patterns [Code]
SGATs	[TKDE 2021] Sparse graph attention networks [Code]
FAGCN	[AAAI 2021] Beyond Low-frequency Information in Graph Convolutional Networks [Code]
Geom-GCN	[ICLR 2019] Geom-GCN: Geometric Graph Convolutional Networks [Code]

Acknowledgement

This page is contributed and maintained by Kaize Ding (kaize.ding@asu.edu), Zhe Xu (zhexu3@illinois.edu).