Research map / Graph learning / Graph algorithms and clustering
Regularized Gaussian Belief Propagation with Nodes of Arbitrary Size
Francois Kamper, Sarel J. Steel, Johan A. du Preez
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 |
|---|---|---|---|
| use higher-dimensional sGaBP-m (cluster nodes) instead of univariate sGaBP for univariate marginals | sGaBP · Simulated Gaussian MG, 100 variables, 10 clusters of 10, 1000 runs | mean KL divergence of univariate marginals: about 32 times more accurate on average | Improved |
| use sGaBP-m instead of RGaBP | RGaBP · Simulated Gaussian MG, 100x100 precision, 10 clusters | mean KL divergence of cluster marginals | Improved |
| use sGaBP-m instead of CFGaBP | CFGaBP · Simulated Gaussian MG, 100x100 precision, 10 clusters | number of iterations to convergence | Improved |
| use sGaBP-m instead of RGaBP | RGaBP · Simulated Gaussian MG, 100x100 precision, 10 clusters | number of iterations to convergence | Improved |
| add node regularization (lambda > 0) to GaBP-m (ablation) | GaBP-m · Simulated Gaussian MG | number of iterations to convergence | Improved |
| set heuristic step size alpha to 0.05 or 0.1 instead of optimal regularization (ablation) | heuristic regularization · Simulated Gaussian MG, 1000 runs | number of iterations to convergence | Mixed |
| set heuristic alpha to 0.01 instead of 0.05 or 0.1 (ablation) | heuristic regularization · Simulated Gaussian MG, 1000 runs | number of iterations to convergence | Hurt |
Closest papers by meaning
- On the Convergence of Gaussian Belief Propagation with Nodes of Arbitrary Size (JMLR 2019 · similarity 0.94)
- Convergence of Gaussian Belief Propagation Under General Pairwise Factorization: Connecting Gaussian MRF with Pairwise Linear Gaussian Model (JMLR 2019 · similarity 0.89)
- Belief Propagation Converges to Gaussian Distributions in Sparsely-Connected Factor Graphs (ICML 2026 · similarity 0.89)
- Turbocharging Gaussian Process Inference with Approximate Sketch-and-Project (NeurIPS 2025 · similarity 0.84)
- Learning Gaussian DAG Models without Condition Number Bounds (ICML 2025 · similarity 0.84)
- Learning Local Neighborhoods of Non-Gaussian Graphical Models (AAAI 2025 · similarity 0.83)
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.
