Research map / Graph learning / Graph algorithms and clustering
Learning Laplacian Matrix from Graph Signals with Sparse Spectral Representation
Pierre Humbert, Batiste Le Bars, Laurent Oudre, Argyris Kalogeratos, Nicolas Vayatis
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 |
|---|---|---|---|
| add spectral sparsity prior (IGL-3SR vs smoothness-only GL-SigRep) | graph learning · RG graphs N=50 | F1-measure: 0.750 -> 0.901 | Improved |
| add spectral sparsity prior (IGL-3SR vs smoothness-only GL-SigRep) | graph learning · ER graphs N=50 | F1-measure: 0.610 -> 0.815 | Improved |
| use relaxed fast FGL-3SR instead of IGL-3SR | 3SR graph learning · RG graphs N=50 | F1-measure: 0.901 -> 0.812 | Hurt |
| use relaxed fast FGL-3SR instead of IGL-3SR | 3SR graph learning · RG graphs N=20 | F1-measure: 0.974 -> 0.968 | No clear effect |
| use FGL-3SR instead of ESA-GL | graph learning · RG graphs N=50 | F1-measure: 0.868 -> 0.812 | Hurt |
| use FGL-3SR runtime instead of IGL-3SR | 3SR graph learning · RG graphs N=100 | Time: < 50mins -> < 2mins | Improved |
| vary alpha in FGL-3SR (ablation) | FGL-3SR · RG N=20, 10-bandlimited | F1-measure | No clear effect |
| increase beta (too much sparsity) (ablation) | FGL-3SR · RG N=20, 10-bandlimited | F1-measure | Hurt |
| increase alpha to large values (ablation) | FGL-3SR · RG N=20, 10-bandlimited | convergence | Hurt |
| increase barrier parameter t (ablation) | IGL-3SR · RG N=20 | learned graph sparsity: t from 10 to 10^4 | Improved |
| use FGL-3SR vs CLR clustering | graph learning · TCGA cancer RNA-Seq | clustering accuracy: 0.9862 -> 0.9887 | No clear effect |
| use SGL instead of FGL-3SR | graph learning · TCGA cancer RNA-Seq | clustering accuracy: 0.9887 -> 0.9987 | Hurt |
Closest papers by meaning
- Nonlinear Laplacians: Tunable principal component analysis under directional prior information (NeurIPS 2025 · similarity 0.89)
- WaveNet: Tackling Non-stationary Graph Signals via Graph Spectral Wavelets (AAAI 2024 · similarity 0.88)
- Taxonomy of reduction matrices for Graph Coarsening (NeurIPS 2025 · similarity 0.87)
- A General Graph Spectral Wavelet Convolution via Chebyshev Order Decomposition (ICML 2025 · similarity 0.87)
- Position: Spectral GNNs Rely Less on Graph Fourier Basis than Conceived (ICML 2025 · similarity 0.86)
- A Poincaré Inequality and Consistency Results for Signal Sampling on Large Graphs (ICLR 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.
