Research map / Graph learning / Graph self-supervised learning
Few-Shot Learning with Graph Neural Networks
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 |
|---|---|---|---|
| aggregate information among nodes with the full GNN instead of metric learning + KNN (ablation) | Our metric learning + KNN · Mini-Imagenet, 5-Way 5-shot | accuracy: 64.02% ±0.51% -> 66.41% ±0.63% | Helped |
| use unlabeled samples (ablation) | GNN · Omniglot, 5-Way 5-shot, 20%-labeled | accuracy: 99.18% -> 99.59% | Helped |
| use unlabeled samples (ablation) | GNN · Omniglot, 5-Way 5-shot, 40%-labeled | accuracy: 99.59% -> 99.63% | No clear effect |
| use unlabeled samples (ablation) | GNN · Mini-Imagenet, 5-Way 5-shot, 20%-labeled | accuracy: 50.33% ±0.36% -> 52.45% ±0.88% | Helped |
| use unlabeled samples (ablation) | GNN · Mini-Imagenet, 5-Way 5-shot, 40%-labeled | accuracy: 56.91% ±0.42% -> 58.76% ±0.86% | Helped |
| use learned active selection instead of random selection (ablation) | GNN active learning · Omniglot, 5-Way 5-shot, 20%-labeled | accuracy: 99.59% -> 99.62% | No clear effect |
| use learned active selection instead of random selection (ablation) | GNN active learning · Mini-Imagenet, 5-Way 5-shot, 20%-labeled | accuracy: 52.56% ±1.18% -> 55.99% ±1.35% | Helped |
Closest papers by meaning
- FEW-SHOT LEARNING ON GRAPHS VIA SUPER-CLASSES BASED ON GRAPH SPECTRAL MEASURES (ICLR 2020 · similarity 0.91)
- Graph Few-Shot Learning via Knowledge Transfer (AAAI 2020 · similarity 0.91)
- Knowledge Graph Transfer Network for Few-Shot Recognition (AAAI 2020 · similarity 0.90)
- Unlocking the Potential of Black-box Pre-trained GNNs for Graph Few-shot Learning (AAAI 2025 · similarity 0.90)
- Hybrid Graph Neural Networks for Few-Shot Learning (AAAI 2022 · similarity 0.90)
- Graph Few-Shot Learning via Adaptive Spectrum Experts and Cross-Set Distribution Calibration (NeurIPS 2025 · similarity 0.88)
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.