Research map / Graph learning / Graph algorithms and clustering
Regularization of Mixture Models for Robust Principal Graph Learning
Aurélien Decelle, Tony Bonnaire, N. Aghanim
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 an explicit uniform background component and learned local variances to graph-regularized mixture model | principal graph learning · Synthetic three-branch dataset with 25% background noise | qualitative graph recovery: estimated background level 25% | Improved |
| increase lambda_sigma toward a large value (ablation) | GRMM · Synthetic three-branch dataset | variance adaptivity | Hurt |
| vary lambda_pi from 0 to 10 (ablation) | GRMM · Synthetic three-branch dataset | graph structure: 0 to 10 | No clear effect |
| add high-probability edges from averaged sub-sampled MSTs to the MST (ablation) | MST graph prior · Synthetic Voronoi dataset | cycles recovered: 27 new edges | Improved |
| use average MST graph instead of persistent homology skeleton | cycle graph construction · Voronoi dataset with outliers | spurious loops and small cycles | Improved |
| add average graph prior to MST on galaxy slice (ablation) | GRMM · IllustrisTNG 2D galaxy slice (N=31500) | cycles represented: 132 edges added | Improved |
Closest papers by meaning
- Graph Convolution for Semi-Supervised Classification: Improved Linear Separability and Out-of-Distribution Generalization (ICML 2021 · similarity 0.86)
- Geometric structure of graph Laplacian embeddings (JMLR 2021 · similarity 0.86)
- Robust Spectral Detection of Global Structures in the Data by Learning a Regularization (NeurIPS 2016 · similarity 0.85)
- PRIMAL-GMM: PaRametrIc MAnifold Learning of Gaussian Mixture Models (TPAMI 2021 · similarity 0.85)
- Recovering Manifold Structure Using Ollivier Ricci Curvature (ICLR 2025 · similarity 0.84)
- Manifold Random Features (NeurIPS 2026 · similarity 0.84)
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.
