Research map / Graph learning / GNN architectures
Adaptive Kernel Graph Neural Network
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 adaptive kernel learning (ablation) | AKGNN · Cora | accuracy: 81.4$\pm$0.2 -> 84.4$\pm$0.3 | Helped |
| use adaptive kernel learning (ablation) | AKGNN · Citeseer | accuracy: 71.9$\pm$0.1 -> 73.5$\pm$0.2 | Helped |
| use adaptive kernel learning (ablation) | AKGNN · Pubmed | accuracy: 79.1$\pm$0.2 -> 80.4$\pm$0.3 | Helped |
| use the parameterization trick (ablation) | AKGNN · Cora | accuracy: 83.1$\pm$0.1 -> 84.4$\pm$0.3 | Helped |
| use the parameterization trick (ablation) | AKGNN · Citeseer | accuracy: 72.2$\pm$0.5 -> 73.5$\pm$0.2 | Helped |
| use the parameterization trick (ablation) | AKGNN · Pubmed | accuracy: 80.1$\pm$0.3 -> 80.4$\pm$0.3 | Helped |
| use sum readout across propagation layers (ablation) | AKGNN · Cora | accuracy: 83.5$\pm$0.2 -> 84.4$\pm$0.3 | Helped |
| use sum readout across propagation layers (ablation) | AKGNN · Citeseer | accuracy: 73.1$\pm$0.3 -> 73.5$\pm$0.2 | Helped |
| use sum readout across propagation layers (ablation) | AKGNN · Pubmed | accuracy: 79.4$\pm$0.2 -> 80.4$\pm$0.3 | Helped |
| replace the strongest reported baseline, APPNP, with AKGNN | node classification · Cora | accuracy: 83.8$\pm$0.3 -> 84.4$\pm$0.3 | Helped |
| replace the strongest reported baseline, GAugO, with AKGNN | node classification · Citeseer | accuracy: 73.3$\pm$1.1 -> 73.5$\pm$0.2 | Helped |
| replace the strongest reported baseline, SSGC, with AKGNN | node classification · Pubmed | accuracy: 80.2$\pm$0.0 -> 80.4$\pm$0.3 | Helped |
Closest papers by meaning
- How Graph Neural Networks Learn: Lessons from Training Dynamics (ICML 2024 · similarity 0.92)
- Learning to Approximate Adaptive Kernel Convolution on Graphs (AAAI 2024 · similarity 0.91)
- Exploring Consistency in Graph Representations: from Graph Kernels to Graph Neural Networks (NeurIPS 2024 · similarity 0.89)
- Graph Neural Networks Are More Than Filters: Revisiting and Benchmarking from A Spectral Perspective (ICLR 2025 · similarity 0.88)
- A Manifold Perspective on the Statistical Generalization of Graph Neural Networks (ICML 2025 · similarity 0.88)
- Spatio-Spectral Graph Neural Networks (NeurIPS 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.