Research map / Graph learning / Graph self-supervised learning
Collaborative Graph Convolutional Networks: Unsupervised Learning Meets Semi-Supervised Learning
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 |
|---|---|---|---|
| use GMM prior instead of single Gaussian prior with VGAE | GMM-VGAE · Cora clustering | Acc: 55.95 -> 71.50 | Helped |
| use clustering-consistent pseudo-labels (CGCN) vs plain self-training (ablation) | GCN · Cora 0.5% labels | accuracy: 53.7 -> 64.3 | Helped |
| joint collaborative training vs clustering pre-training then training classifier (Two-stage) (ablation) | CGCN · Cora 0.5% labels | accuracy: 57.9 -> 64.3 | Helped |
| joint collaborative training vs Two-stage (ablation) | CGCN · Cora 5% labels | accuracy: 83.3 -> 84.2 | Helped |
| joint collaborative training vs Two-stage (ablation) | CGCN · CiteSeer 0.5% labels | accuracy: 51.3 -> 59.3 | Helped |
| joint collaborative training vs Two-stage (ablation) | CGCN · PubMed 0.3% labels | accuracy: 78.2 -> 80.3 | Helped |
| consistency of clustering and GCN pseudo-labels (CGCN) vs union of random-walk and GCN labels (ablation) | CGCN · Cora 1% labels | accuracy: 69.9 -> 72.4 | Helped |
| train GMM-VGAE collaboratively in CGCN with shared first layer (ablation) | GMM-VGAE · Cora clustering accuracy | accuracy: 71.5 -> 73.8 | Helped |
| train GMM-VGAE collaboratively in CGCN with shared first layer, at high label rates (ablation) | GMM-VGAE · CiteSeer clustering accuracy | accuracy: 67.4 -> 70.2 | Mixed |
| CGCN vs GCN with validation at low labels on Cora | GCN · Cora 0.5% labels | accuracy: 50.9 -> 64.3 | Helped |
Closest papers by meaning
- Deeper Insights Into Graph Convolutional Networks for Semi-Supervised Learning (AAAI 2018 · similarity 0.92)
- Graph Convolutional Networks Meet Markov Random Fields: Semi-Supervised Community Detection in Attribute Networks (AAAI 2019 · similarity 0.92)
- Semi-Supervised Classification with Graph Convolutional Networks (ICLR 2017 · similarity 0.91)
- Semi-Supervised Learning With Graph Learning-Convolutional Networks (CVPR 2019 · similarity 0.91)
- GC-Flow: A Graph-Based Flow Network for Effective Clustering (ICML 2023 · similarity 0.90)
- Unsupervised Neighborhood Propagation Kernel Layers for Semi-supervised Node Classification (AAAI 2024 · similarity 0.89)
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.