Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Decomposition for Bayesian Networks: Local and Parallel Inference
Pei Heng, Xinyi Hu, Yi Sun
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 minimal d-decomposition tree pruning instead of junction tree | junction-tree inference · Pathfinder joint query, |Q|=2 | inference time: 0.7743 -> 0.0613 | Improved |
| use pruning instead of lossless decomposition | lossless decomposition · Pathfinder joint query, |Q|=2 | inference time: 0.5061 -> 0.0613 | Improved |
| use pruning instead of junction tree at |Q|=10 | junction-tree inference · Pathfinder joint query, |Q|=10 | inference time: 0.7744 -> 0.1855 | Improved |
| use pruning instead of junction tree at |Q|=20 | junction-tree inference · Pathfinder joint query, |Q|=20 | inference time: 0.8450 -> 0.2966 | Improved |
| use pruning instead of junction tree at |Q|=25 | junction-tree inference · Pathfinder joint query, |Q|=25 | inference time: 1.8515 -> 1.8073 | No clear effect |
| use pruning instead of lossless decomposition at |Q|=25 | lossless decomposition · Pathfinder joint query, |Q|=25 | inference time: 30.1472 -> 1.8073 | Improved |
| increase query dimension from 2 to 25 | minimal d-decomposition tree inference · Pathfinder joint query | inference time: 0.0613 -> 1.8073 | Hurt |
| parallel parameter learning on decomposed sub-models instead of global learning | global parameter estimation · Child (small network) | speed-up ratio: 1.40 | Mixed |
| parallel parameter learning on decomposed sub-models instead of global learning | global parameter estimation · Pigs (large network) | speed-up ratio: 2.69 | Improved |
| increase sample size to 10,000 | parallel decomposition learning · Multinomial BNs | KL, Hellinger and Bhattacharyya distance | Improved |
| increase sample size to 10,000 | parallel decomposition learning · Gaussian BNs | Wasserstein-2 distance: 0.04 | Improved |
Closest papers by meaning
- Parallel Bayesian Network Structure Learning (ICML 2018 · similarity 0.88)
- Learning Bayesian Networks with Bounded Tree-width via Guided Search (AAAI 2016 · similarity 0.88)
- Learning Fast-Inference Bayesian Networks (NeurIPS 2021 · similarity 0.88)
- Learning Treewidth-Bounded Bayesian Networks with Thousands of Variables (NeurIPS 2016 · similarity 0.87)
- Learning Big Gaussian Bayesian Networks: Partition, Estimation and Fusion (JMLR 2020 · similarity 0.86)
- On the Expressive Power of Tree-Structured Probabilistic Circuits (NeurIPS 2024 · similarity 0.85)
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.
