Research map / Graph learning / Graph self-supervised learning
Graph Contrastive Learning with Augmentations
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| GraphCL pre-training | GNN, semi-supervised 1% labels · MNIST superpixel | accuracy: 60.39 -> 83.41 | Helped |
| GraphCL pre-training | GNN, transfer learning · MoleculeNet / PPI | ROC-AUC: MUV 71.8 -> 69.80; HIV 75.3 -> 78.47 | Mixed |
| graph contrastive learning with no augmentation (ablation) | GNN · NCI1/PROTEINS/COLLAB/RDT-B | accuracy | Hurt |
| add appropriate single augmentation (ablation) | GraphCL · COLLAB | accuracy: 6.27% | Helped |
| compose two different augmentation types (ablation) | GraphCL · COLLAB | accuracy: 7.11% | Helped |
| edge perturbation augmentation (ablation) | GraphCL · NCI1 | accuracy | Hurt |
| edge perturbation augmentation (ablation) | GraphCL · COLLAB, RDT-B | accuracy | Helped |
| attribute masking (ablation) | GraphCL · COLLAB vs RDT-B | accuracy: 5.12% vs 0.17% | Mixed |
| GraphCL pre-training | GNN 4-layer · RL-S2V adversarial attack | accuracy: 70.20 -> 84.86 | Helped |
Closest papers by meaning
- Graph Contrastive Learning Automated (ICML 2021 · similarity 0.94)
- Community-Invariant Graph Contrastive Learning (ICML 2024 · similarity 0.92)
- Adversarial Contrastive Graph Augmentation with Counterfactual Regularization (AAAI 2025 · similarity 0.91)
- ReGCL: Rethinking Message Passing in Graph Contrastive Learning (AAAI 2024 · similarity 0.91)
- Unified Graph Augmentations for Generalized Contrastive Learning on Graphs (NeurIPS 2024 · similarity 0.91)
- Perfect Alignment May be Poisonous to Graph Contrastive Learning (ICML 2024 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.