Research map / Graph learning / Graph algorithms and clustering
Community-Based Group Graphical Lasso
Eugen Pircalabelu, Gerda Claeskens
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 |
|---|---|---|---|
| increase sample size from 100 to 500 to 1000 | ComGGL2 · Simulated Gaussian (Kn=1, 20 nodes/community) | Frobenius norm error: 5.41 -> 4.74 -> 4.69 | Improved |
| replace Gaussian data with Student t2 (misspecification) | ComGGL2 · Simulated (Kn=1, 20 nodes/community, n=100) | Frobenius norm error: 5.41 -> 8.13 | Hurt |
| use group penalty with estimated communities instead of plain graphical lasso plus clustering | ComGGL1 · rsfMRI (Kn=7) | F1 label accuracy: 0.50 vs 0.37 | Improved |
| enforce positive definiteness (ComGGL2 vs ComGGL1) (ablation) | ComGGL1 · rsfMRI lobe homogeneity | average homogeneity over lobes: 7.24 -> 6.66 | Hurt |
| use ComGGL instead of two-step approaches on simulated graphs | ComGGL2 · Simulated Gaussian (Kn=3, 20 nodes/community, n=500) | Frobenius norm error: 6.09 vs 6.00 | No clear effect |
| use average linkage clusterGL instead of single | clusterGL · Simulated Gaussian (Kn=3, 20 nodes/community, n=500) | F1 edge recovery: 0.63 -> 0.41 | Hurt |
| use complete linkage clusterGL instead of single | clusterGL · Simulated Gaussian (Kn=3, 20 nodes/community, n=500) | Frobenius norm error: 6.40 -> 8.71 | Hurt |
| increase number of communities Kn above 4 | clusterGL · rsfMRI (F1) | F1 label accuracy: 0.37 -> 0.25 -> 0.17 | Hurt |
| use ComGGL1 instead of CORD for lobe homogeneity | ComGGL1 · rsfMRI Occipital lobe | homogeneity: 13.0 vs 4.33 | Improved |
Closest papers by meaning
- Simultaneous Identification of Sparse Structures and Communities in Heterogeneous Graphical Models (JMLR 2026 · similarity 0.91)
- High-Dimensional Inference for Cluster-Based Graphical Models (JMLR 2020 · similarity 0.88)
- Equivalence of Graphical Lasso and Thresholding for Sparse Graphs (JMLR 2016 · similarity 0.87)
- Community detection in sparse latent space models (JMLR 2022 · similarity 0.87)
- Bayesian Joint Estimation of Multiple Graphical Models (NeurIPS 2019 · similarity 0.87)
- vGraph: A Generative Model for Joint Community Detection and Node Representation Learning (NeurIPS 2019 · similarity 0.87)
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.
