Research map / Graph learning / Graph self-supervised learning
Constant Curvature Graph Convolutional Networks
Gregor Bachmann, Gary Becigneul, Octavian Ganea
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 hyperbolic κ-GCN for tree embeddings instead of the strongest Euclidean baseline | Graph metric embedding · Tree | average distortion: 0.045 -> 0.0029 | Improved |
| Use product spherical κ-GCN for toroidal embeddings instead of the strongest Euclidean baseline | Graph metric embedding · Toroidal | average distortion: 0.0603 -> 0.0464 | Improved |
| Use spherical κ-GCN for spherical graph embeddings instead of the strongest Euclidean baseline | Graph metric embedding · Spherical Graph | average distortion: 0.0409 -> 0.0337 | Improved |
| Use hyperbolic κ-GCN instead of Euclidean GCN | Node classification · Citeseer | accuracy: 0.729 ± 0.0054 -> 0.732 ± 0.0051 | Improved |
| Use spherical κ-GCN instead of Euclidean GCN | Node classification · Cora | accuracy: 0.814 ± 0.004 -> 0.819 ± 0.0045 | Improved |
| Use spherical κ-GCN instead of the strongest hyperbolic baseline | Node classification · Pubmed | accuracy: 0.798 ± 0.0043 -> 0.788 ± 0.0049 | Hurt |
| Use hyperbolic κ-GCN instead of the strongest hyperbolic baseline | Node classification · Airport | accuracy: 0.844 ± 0.0041 -> 0.819 ± 0.0033 | Hurt |
| Match the embedding geometry to the graph's underlying geometry | κ-GCN graph embeddings · Tree, Toroidal, and Spherical Graph | average distortion | Improved |
Closest papers by meaning
- Pseudo-Riemannian Graph Convolutional Networks (NeurIPS 2022 · similarity 0.92)
- Curvature Graph Network (ICLR 2020 · similarity 0.90)
- Spectro-Riemannian Graph Neural Networks (ICLR 2025 · similarity 0.88)
- Building Transformation Layers for Riemannian Neural Networks (NeurIPS 2026 · similarity 0.88)
- Effective Structural Encodings via Local Curvature Profiles (ICLR 2024 · similarity 0.87)
- Graph Learning in 4D: A Quaternion-Valued Laplacian to Enhance Spectral GCNs (AAAI 2024 · similarity 0.86)
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.
