Research map / Learning theory and algorithms / Combinatorial and search optimization
Belief Propagation Neural Networks
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 |
|---|---|---|---|
| replace standard BP updates with learned BPNN-D updates | BP on Ising models · Ising models, validation (44 models where BP converged) | RMSE of log partition function: 7.20 -> .97 | Helped |
| replace standard BP updates with learned BPNN-D updates | BP on Ising models · Ising models, validation | convergence speed: 1.7x | Helped |
| replace BP with BPNN | partition-function estimation · Stochastic Block Model, Validation | RMSE of log partition function: 11.14 -> 4.15 | Helped |
| replace GNN with BPNN | partition-function estimation · Stochastic Block Model, Validation | RMSE of log partition function: 7.93 -> 4.15 | Helped |
| add the no-double-counting message update (ablation) | BPNN · Stochastic Block Model, Validation | RMSE of log partition function: 8.43 -> 4.15 | Helped |
| add local variable permutation invariance (ablation) | BPNN · Stochastic Block Model, Validation | RMSE of log partition function: 5.63 -> 4.15 | Helped |
| replace F2 with BPNN | model counting · or_50 validation | RMSE of log model count: 2.5 -> .30 | Helped |
| use BPNN-P on GPU instead of F2 | model counting · or_50 benchmarks | median speedup: 248 | Helped |
| use BPNN-P on GPU instead of ApproxMC3 | model counting · or_50 benchmarks | median speedup: 3,689 | Helped |
Closest papers by meaning
- Deep Attentive Belief Propagation: Integrating Reasoning and Learning for Solving Constraint Optimization Problems (NeurIPS 2022 · similarity 0.88)
- Are Graph Neural Networks Optimal Approximation Algorithms? (NeurIPS 2024 · similarity 0.86)
- Belief Propagation Reloaded: Learning BP-Layers for Labeling Problems (CVPR 2020 · similarity 0.86)
- COMBHelper: A Neural Approach to Reduce Search Space for Graph Combinatorial Problems (AAAI 2024 · similarity 0.84)
- Scalable Neural Network Verification with Branch-and-bound Inferred Cutting Planes (NeurIPS 2024 · similarity 0.84)
- BIPNN: Learning to Solve Binary Integer Programming via Hypergraph Neural Networks (NeurIPS 2025 · 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.