Research map / Graph learning / Graph algorithms and clustering
Hunt For The Unique, Stable, Sparse And Fast Feature Learning On Graphs
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 FGSD (harmonic distance histogram features, SVM) instead of baselines for unlabeled graph classification | graph classification · unlabeled bioinformatics datasets (Table 2) | classification accuracy | Helped |
| Use FGSD features instead of PSCN on social network datasets | graph classification · social network datasets (Table 3) | classification accuracy | Mixed |
| Use FGSD (no node labels) instead of labeled methods on labeled bioinformatics datasets | graph classification · labeled bioinformatics datasets (Table 4) | classification accuracy | Mixed |
| Choose harmonic (f(λ)=1/λ) rather than other f(λ) for sparsity of features | FGSD feature histogram · MUTAG feature matrix | matrix sparsity | Helped |
| Use biharmonic instead of harmonic distance, favoring uniqueness over sparsity | FGSD feature histogram · preliminary experiments on graph classification (supplementary) | classification accuracy | Hurt |
| Use FGSD exact computation (sparse Cholesky-based pseudoinverse) for runtime | FGSD · unlabeled datasets wall-time | runtime: 2x-1000x faster than others | Helped |
Closest papers by meaning
- Rethinking Kernel Methods for Node Representation Learning on Graphs (NeurIPS 2019 · similarity 0.88)
- MMD Graph Kernel: Effective Metric Learning for Graphs via Maximum Mean Discrepancy (ICLR 2024 · similarity 0.87)
- Graph Filtration Kernels (AAAI 2022 · similarity 0.87)
- SpeAr: A Spectral Approach for Zero-Shot Node Classification (NeurIPS 2024 · similarity 0.86)
- Spectral Graph Sparsification Preserves Representation Geometry in Graph Neural Networks (NeurIPS 2026 · similarity 0.86)
- Subgraph Invariant Learning Towards Large-Scale Graph Node Classification (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.