Augmentation-Free Self-Supervised Learning on Graphs
About
Inspired by the recent success of self-supervised methods applied on images, self-supervised learning on graph structured data has seen rapid growth especially centered on augmentation-based contrastive methods. However, we argue that without carefully designed augmentation techniques, augmentations on graphs may behave arbitrarily in that the underlying semantics of graphs can drastically change. As a consequence, the performance of existing augmentation-based methods is highly dependent on the choice of augmentation scheme, i.e., hyperparameters associated with augmentations. In this paper, we propose a novel augmentation-free self-supervised learning framework for graphs, named AFGRL. Specifically, we generate an alternative view of a graph by discovering nodes that share the local structural information and the global semantics with the graph. Extensive experiments towards various node-level tasks, i.e., node classification, clustering, and similarity search on various real-world datasets demonstrate the superiority of AFGRL. The source code for AFGRL is available at https://github.com/Namkyeong/AFGRL.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Node Classification | Citeseer | Accuracy68.7 | 931 | |
| Node Classification | Pubmed | Accuracy79.71 | 819 | |
| Node Classification | Photo | Mean Accuracy93.25 | 343 | |
| Node Classification | Pubmed | Accuracy79.71 | 178 | |
| Molecular property prediction | MoleculeNet BBBP (scaffold) | ROC AUC70.8 | 140 | |
| Node Classification | Photo | Accuracy93.25 | 139 | |
| Molecular property prediction | MoleculeNet SIDER (scaffold) | ROC-AUC0.602 | 120 | |
| Molecular property prediction | MoleculeNet BACE (scaffold) | ROC-AUC79.3 | 110 | |
| Molecular property prediction | MoleculeNet MUV (scaffold) | ROC-AUC0.777 | 91 | |
| Node Classification | Computer | Accuracy89.9 | 89 |