Research map / Probabilistic and scientific modeling / Variational and generative sampling
Poisson Sum-Product Networks: A Deep Architecture for Tractable Multivariate Poisson Distributions
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 PSPN instead of Gaussian SPN for count data | PSPN · C&C, MSNBC, NIPS, synthetic count data | prediction error | Helped |
| use PSPN instead of Poisson dependency networks (PDN) | PSPN · C&C, MSNBC, NIPS, synthetic count data | prediction error | No clear effect |
| use deeper PSPN instead of shallow LDA / hierarchical LDA | PSPN · NIPS count data | perplexity | Helped |
| use PSPN instead of PDN for training time | PSPN · count benchmarks | run time | Hurt |
| use randomized PSPN learner (subsampled decomposition, random hyperplane conditioning) (ablation) | PSPN · count benchmarks | perplexity and run time | Mixed |
| change instability-test significance threshold p (1e-5, 1e-4, 1e-3) (ablation) | PSPN decomposition · simulated XMRF networks | edge recovery difference | No clear effect |
| use PSPN decomposition instead of LPGM for network discovery | PSPN · simulated XMRF networks | edge recovery difference | Helped |
| use PSPN per class instead of SVM on Weisfeiler-Lehman features | PSPN · Mutag, PTC, Enzymes | accuracy | No clear effect |
Closest papers by meaning
- Sum-Product-Set Networks: Deep Tractable Models for Tree-Structured Graphs (ICLR 2024 · similarity 0.86)
- Causal Discovery from Poisson Branching Structural Causal Model Using High-Order Cumulant with Path Analysis (AAAI 2024 · similarity 0.85)
- On the Identifiability of Poisson Branching Structural Causal Model Using Probability Generating Function (NeurIPS 2024 · similarity 0.85)
- Fully Heteroscedastic Count Regression with Deep Double Poisson Networks (ICML 2025 · similarity 0.85)
- Distribution Learning Meets Graph Structure Sampling (NeurIPS 2025 · similarity 0.84)
- Data subsampling for Poisson regression with pth-root-link (NeurIPS 2024 · 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.