Research map / Probabilistic and scientific modeling / Variational and generative sampling
Maximum A Posteriori Inference in Sum-Product 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 |
|---|---|---|---|
| Use KBT100 instead of existing approximate solvers | MAP inference in LearnSPN SPNs · 20 real-world datasets, Q/E/H 3/3/4 | winning count | Mixed |
| Increase K from 10 to 100 in KBT (ablation) | KBT MAP solver · 20 real-world datasets | winning count | Helped |
| Increase K from 10 to 100 in KBT (ablation) | KBT MAP solver · 20 real-world datasets | running time | Hurt |
| Add Forward Checking instead of Marginal Checking (ablation) | exact MAP solver · real-world datasets, Q/E/H 3/3/4 | finishing count within 10 minutes | Helped |
| Add variable/value Ordering to Forward Checking (ablation) | exact MAP solver · real-world datasets, Q/E/H 3/3/4 | finishing count within 10 minutes | Helped |
| Add Stage to FC with Ordering (ablation) | exact MAP solver · real-world datasets, Q/E/H 3/3/4 | finishing count within 10 minutes | Helped |
| Use Ordering to guide the search (ablation) | anytime exact MAP solver · real-world datasets, Q/E/H 3/3/4 | winning count | Helped |
| Increase the number of hidden variables | approximate MAP solvers · real-world datasets with varying Q/E/H proportions | winning count | Mixed |
Closest papers by meaning
- Collapsed Variational Inference for Sum-Product Networks (ICML 2016 · similarity 0.92)
- Probabilistic Circuits for Variational Inference in Discrete Graphical Models (NeurIPS 2020 · similarity 0.90)
- Neural Network Approximators for Marginal MAP in Probabilistic Circuits (AAAI 2024 · similarity 0.90)
- Sum-Product-Set Networks: Deep Tractable Models for Tree-Structured Graphs (ICLR 2024 · similarity 0.88)
- Tractable Probabilistic Graph Representation Learning with Graph-Induced Sum-Product Networks (ICLR 2024 · similarity 0.87)
- Inferring Dynamic Networks from Marginals with Iterative Proportional Fitting (ICML 2024 · similarity 0.86)
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.