Research map / Graph learning / Graph self-supervised learning
Label Efficient Semi-Supervised Learning via Graph Filtering
Check 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 |
|---|---|---|---|
| Use filtered features and a supervised classifier (GLP) instead of feature-only MLP | Cora, 4 labels per class · Cora, 4 labels per class | accuracy: 36.4 -> 68.0 | Helped |
| Use filtered features and a supervised classifier (GLP) instead of feature-only MLP | CiteSeer, 4 labels per class · CiteSeer, 4 labels per class | accuracy: 38.0 -> 57.3 | Helped |
| Use IGCN with an adjustable graph filter instead of GCN | Cora, 4 labels per class · Cora, 4 labels per class | accuracy: 65.2 -> 70.3 | Helped |
| Use IGCN with an adjustable graph filter instead of GCN | NELL, 1% label rate · NELL, 1% label rate | accuracy: 63.9 -> 76.7 | Helped |
| Increase IGCN RNM filter exponent from k=1 to k=2 (ablation) | IGCN(RNM) · AWA2, unseen classes | accuracy: 77.9 -> 77.7 | Hurt |
| Increase IGCN RNM filter exponent from k=1 to k=3 (ablation) | IGCN(RNM) · AWA2, unseen classes | accuracy: 77.9 -> 73.1 | Hurt |
| Increase GLP RNM filter exponent from k=2 to k=4 (ablation) | GLP(RNM) · AWA2, unseen classes | accuracy: 76.0 -> 75.0 | Hurt |
| Increase GLP RNM filter exponent from k=2 to k=6 (ablation) | GLP(RNM) · AWA2, unseen classes | accuracy: 76.0 -> 73.0 | Hurt |
| Replace the 6-layer GCN with a 2-layer IGCN using k=3 | GCNZ · AWA2, unseen classes | accuracy: 68.0 -> 73.1 | Helped |
Closest papers by meaning
- A Flexible Generative Framework for Graph-based Semi-supervised Learning (NeurIPS 2019 · similarity 0.90)
- Semi-Supervised Classification with Graph Convolutional Networks (ICLR 2017 · similarity 0.90)
- Shoestring: Graph-Based Semi-Supervised Classification With Severely Limited Labeled Data (CVPR 2020 · similarity 0.90)
- Semi-Supervised Learning With Graph Learning-Convolutional Networks (CVPR 2019 · similarity 0.90)
- HGDL: Heterogeneous Graph Label Distribution Learning (NeurIPS 2024 · similarity 0.88)
- Normalize Then Propagate: Efficient Homophilous Regularization for Few-Shot Semi-Supervised Node Classification (AAAI 2025 · similarity 0.87)
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.