Data Augmentation for Graph Neural Networks

By Tong Zhao, Yozen Liu, Leonardo Neves, Oliver J. Woodford, Meng Jiang, Neil Shah. AAAI Conference on Artificial Intelligence, 2020

507 citations recorded by the source corpus.

Data augmentation has been widely used to improve generalizability of machine learning models. However, comparatively little work studies data augmentation for graphs. This is largely due to the complex, non-Euclidean structure of graphs, which limits possible manipulation operations. Augmentation operations commonly used in vision and language have no analogs for graphs. Our work studies graph data augmentation for graph neural networks (GNNs) in the context of improving semi-supervised node-classification. We discuss practical and theoretical motivations, considerations and strategies for graph data augmentation. Our work shows that neural edge predictors can effectively encode class-homophilic structure to promote intra-class edges and demote inter-class edges in given graph structure, and our main contribution introduces the GAug graph data augmentation framework, which leverages these insights to improve performance in GNN-based node classification via edge prediction. Extensive experiments on multiple benchmarks show that augmentation via GAug improves performance across GNN architectures and datasets.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/27d5be9322d71b6fd2faa8a6b87250127a12c0cf

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Yao Ma
2. Tianlong Chen
3. Jiliang Tang
4. Yong Li
5. Yu Zheng
6. Xiaorui Liu
7. Neil Shah
8. Connor Shorten

Citations in the corpus

• Augmentation-Free Self-Supervised Learning on Graphs (2021, background): “…augmentation techniques for graphs have been introduced. e.g., node dropping (You et al. 2020), edge modification (Jin et al. 2021; Qiu et al. 2020; Zhao et al. 2020), subgraph extraction (Jiao et al. 2020; Sun et al. 2021b), attribute masking (Zhu et al. 2020, 2021) and others (Hassani and…”
• Graph Contrastive Learning Automated (2021, background): “For example, the state-of-the-art graph contrastive learning framework, GraphCL (You et al., 2020a), constructs specific contrastive views of graph data via hand-picking ad-hoc augmentations for every dataset (You et al., 2020a; Zhao et al., 2020; Kong et al., 2020).”
• Text Data Augmentation for Deep Learning (2021, background): “Zhao et al. [46] propose an edge augmentation technique that “exposes GNNs to likely (but nonexistent) edges and limiting exposure to unlikely (but existent) ones” [46].”
• Sequential Recommendation with Graph Neural Networks (2021, background): “Common options for kernel-basedmethods include cosine distance [30, 41], Euclidean distance [38, 43] and Mahalanobis distance [18, 34].”
• Graph Neural Network for Traffic Forecasting: A Survey (2021, unknown): “semi-supervised node classification task [251].”
• Is Homophily a Necessity for Graph Neural Networks? (2021, background): “Hence, several recent works claim [41, 40, 4], assume [12, 35, 38] or remark upon [1, 22, 14] GNN models homophily-reliance or unsuitability in capturing heterophily.”