Research map / Graph learning / Graph self-supervised learning
A Flexible Generative Framework for Graph-based 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 the joint generative framework with LSM-GCN instead of GCN | GCN · Cora, standard benchmark | accuracy: 0.815 -> 0.825 | Helped |
| Use the joint generative framework with LSM-GAT instead of GAT | GAT · Cora, standard benchmark | accuracy: 0.825 -> 0.829 | Helped |
| Use the joint generative framework with SBM-GCN instead of GCN | GCN · Pubmed, standard benchmark | accuracy: 0.794 -> 0.784 | Hurt |
| Use the joint generative framework with LSM-GCN instead of GCN | GCN · Citeseer, standard benchmark | accuracy: 0.718 -> 0.744 | Helped |
| Use the joint generative framework with LSM-GAT instead of GAT | GAT · Citeseer, standard benchmark | accuracy: 0.715 -> 0.731 | Helped |
| Use the joint generative framework with LSM-GCN instead of GCN | GCN · Cora, test nodes with edges removed | accuracy: 0.665 -> 0.711 | Helped |
| Use the joint generative framework with LSM-GCN instead of GCN | GCN · Pubmed, test nodes with edges removed | accuracy: 0.746 -> 0.766 | Helped |
| Use the joint generative framework with LSM-GCN instead of GCN | GCN · Citeseer, test nodes with edges removed | accuracy: 0.652 -> 0.704 | Helped |
| Use the joint generative framework with SBM-GAT instead of GAT | GAT · Cora, half the training labels removed per class | accuracy: 0.771 -> 0.796 | Helped |
| Use the joint generative framework with LSM-GCN instead of GCN | GCN · Pubmed, half the training labels removed per class | accuracy: 0.724 -> 0.709 | Hurt |
| Use the joint generative framework with SBM-GCN instead of GCN | GCN · Citeseer, half the training labels removed per class | accuracy: 0.666 -> 0.703 | Helped |
Closest papers by meaning
- Semi-Supervised Classification with Graph Convolutional Networks (ICLR 2017 · similarity 0.91)
- Data driven semi-supervised learning (NeurIPS 2021 · similarity 0.90)
- Label Efficient Semi-Supervised Learning via Graph Filtering (CVPR 2019 · similarity 0.90)
- Graph inference learning for semi-supervised classification (ICLR 2020 · similarity 0.90)
- Contrastive and Generative Graph Convolutional Networks for Graph-based Semi-Supervised Learning (AAAI 2021 · similarity 0.90)
- Hypergraph-enhanced Dual Semi-supervised Graph Classification (ICML 2024 · 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.