Research map / Graph learning / GNN architectures
Bayesian Graph Convolutional Neural Networks for Semi-Supervised Classification
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 |
|---|---|---|---|
| Model graph structure as uncertain using an assortative mixed-membership block model and Bayesian GCNN inference | GCNN node classification · Cora and CiteSeer, limited-label classification | accuracy | Helped |
| Use Bayesian GCNN rather than GCNN | Cora · Selected nodes under random attack | accuracy: 85.55% -> 55.50%; 86.50% -> 69.50% | Helped |
| Use Bayesian GCNN rather than GCNN | CiteSeer · Selected nodes under random attack | accuracy: 88.5% -> 43.0%; 87.0% -> 66.5% | Helped |
| Use Bayesian GCNN rather than GCNN | Cora · Selected nodes under random attack | classifier margin: 0.152 vs 0.387 | Helped |
| Use Bayesian GCNN rather than GCNN | CiteSeer · Selected nodes under random attack | classifier margin: 0.014 vs 0.335 | Helped |
| Use Bayesian GCNN rather than GCNN | Pubmed · 10- and 20-label classification, both splits | accuracy: 73.9±3.4 vs 73.3±3.1; 77.5±2.5 vs 76.0±2.6; 72.8±0.5 vs 72.3±0.8; 78.9±0.3 vs 76.6±0.7 | Hurt |
| Use Bayesian GCNN rather than GCNN | Pubmed · 5-label classification, random and fixed split | accuracy: 69.7±4.5 vs 70.2±4.5; 70.0±0.6 vs 70.9±0.8 | Helped |
Closest papers by meaning
- Deeper Insights Into Graph Convolutional Networks for Semi-Supervised Learning (AAAI 2018 · similarity 0.92)
- Semi-Supervised Classification with Graph Convolutional Networks (ICLR 2017 · similarity 0.91)
- Variational Inference for Graph Convolutional Networks in the Absence of Graph Data and Adversarial Settings (NeurIPS 2020 · similarity 0.91)
- LEARNING GUARANTEES FOR GRAPH CONVOLUTIONAL NETWORKS ON THE STOCHASTIC BLOCK MODEL (ICLR 2022 · similarity 0.90)
- On the Stability of Graph Convolutional Neural Networks: A Probabilistic Perspective (NeurIPS 2025 · similarity 0.90)
- Generalizing CNNs to graphs with learnable neighborhood quantization (NeurIPS 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.