Research map / Graph learning / Graph self-supervised learning
Variational Inference for Graph Convolutional Networks in the Absence of Graph Data and Adversarial Settings
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 |
|---|---|---|---|
| Infer and adjust a posterior over graph edges using a GCN likelihood | VGCN · Citeseer, no graph | test accuracy | Helped |
| Infer and adjust a posterior over graph edges using a GCN likelihood | VGCN · Cora, no graph | test accuracy | Helped |
| Infer a graph posterior to suppress harmful added edges | VGCN · Citeseer and Cora, adversarial graphs with added edges | test accuracy | Helped |
| Infer and adjust the graph posterior | VGCN · Ground-truth Citeseer, Cora, and Polblogs graphs | test accuracy | Helped |
| Replace the free posterior with a low-rank posterior parameterization (ablation) | VGCN · Citation networks | test accuracy | Hurt |
| Replace the free posterior with a low-rank posterior parameterization (ablation) | VGCN · Citation networks | test accuracy variance | Hurt |
| Replace the relaxed Concrete posterior with a discrete Bernoulli posterior (ablation) | VGCN · Citeseer and Cora, no graph | test accuracy | No clear effect |
| Replace the relaxed Concrete posterior with a discrete Bernoulli posterior (ablation) | VGCN · Citeseer and Cora, adversarial | test accuracy | Hurt |
| Use S=3 rather than S=16 posterior samples for prediction (ablation) | VGCN · Prediction | test accuracy | Hurt |
| Use the relaxed Concrete posterior rather than the discrete posterior (ablation) | VGCN · Training convergence | convergence speed | Helped |
Closest papers by meaning
- Bayesian Graph Convolutional Neural Networks for Semi-Supervised Classification (AAAI 2019 · similarity 0.91)
- On the Stability of Graph Convolutional Neural Networks: A Probabilistic Perspective (NeurIPS 2025 · similarity 0.90)
- Convergence and Stability of Graph Convolutional Networks on Large Random Graphs (NeurIPS 2020 · similarity 0.89)
- Generalizing CNNs to graphs with learnable neighborhood quantization (NeurIPS 2024 · similarity 0.87)
- A Manifold Perspective on the Statistical Generalization of Graph Neural Networks (ICML 2025 · similarity 0.87)
- TRACI: A Data-centric Approach for Multi-Domain Generalization on Graphs (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.