Research map / Graph learning / Graph self-supervised learning
Curvature Regularization to Prevent Distortion in Graph Embedding
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 |
|---|---|---|---|
| add curvature regularization (sampled, Omega_s) | MF · Cora node classification | accuracy (%): 44.30 -> 48.37 | Helped |
| add curvature regularization (Omega_c) | Laplacian Eigenmaps · Cora node classification | accuracy (%): 44.30 -> 45.80 | Helped |
| add curvature regularization (Omega_c) | SDNE · Cora node classification | accuracy (%): 42.07 -> 46.40 | Helped |
| add approximated curvature regularization (Omega_a) | DeepWalk · Cora node classification | accuracy (%): 54.65 -> 62.05 | Helped |
| add approximated curvature regularization (Omega_a) | Node2vec · Cora node classification | accuracy (%): 48.02 -> 52.91 | Helped |
| add approximated curvature regularization (Omega_a) | DeepWalk · Citeseer node classification | accuracy (%): 42.24 -> 53.42 | Helped |
| add curvature regularization (Omega_c) | SDNE · Polblogs node classification | accuracy (%): 78.13 -> 76.24 | Hurt |
| add curvature regularization (Omega_c) | MF · Cora link prediction | mean average precision (%): 47.73 -> 49.83 | Helped |
| add curvature regularization (Omega_c) | Laplacian Eigenmaps · Cora link prediction | mean average precision (%): 49.82 -> 54.18 | Helped |
| add curvature regularization (Omega_c) | SDNE · Cora link prediction | mean average precision (%): 74.65 -> 78.78 | Helped |
| add approximated curvature regularization (Omega_a) | DeepWalk · Citeseer link prediction | mean average precision (%): 57.84 -> 72.97 | Helped |
| add approximated curvature regularization (Omega_a) | Node2vec · Cora link prediction | mean average precision (%): 66.18 -> 72.05 | Helped |
Closest papers by meaning
- Graph Geometry-Preserving Autoencoders (ICML 2024 · similarity 0.89)
- Euclidean Embedding of Data Using Local Distances (NeurIPS 2026 · similarity 0.88)
- Toward a Unified Geometry Understanding : Riemannian Diffusion Framework for Graph Generation and Prediction (NeurIPS 2025 · similarity 0.87)
- Recovering Manifold Structure Using Ollivier Ricci Curvature (ICLR 2025 · similarity 0.87)
- Effective Structural Encodings via Local Curvature Profiles (ICLR 2024 · similarity 0.86)
- Weighted Embeddings for Low-Dimensional Graph Representation (AAAI 2025 · 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.