Research map / Probabilistic and scientific modeling / Molecules and proteins
Hierarchical Compound Poisson Factorization
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 |
|---|---|---|---|
| Decouple the response distribution from the Poisson sparsity process using a selectable element distribution | HCPF versus HPF · Ratings, social activity, and music data sets | test log likelihood per thousand entries: amazon: -1.453 -> -0.387; movielens: -76.182 -> -27.056; netflix: -88.490 -> -50.152; yelp: -2.026 -> -0.625; wordpress: | Helped |
| Use HCPF rather than HPF | HCPF versus HPF · merck | test log likelihood per thousand entries: -86.492 -> -93.989 | Hurt |
| Use HCPF rather than HPF | HCPF versus HPF · bestbuy | test log likelihood per thousand entries: -0.289 -> -0.215 | No clear effect |
| Model full-matrix sparsity and responses jointly with HCPF-BI instead of HPF trained on non-missing entries | HCPF-BI versus HPF-NM · amazon, netflix, and yelp | non-missing test log likelihood: amazon: -1.868 -> -1.559; netflix: -1.672 -> -1.586; yelp: -1.940 -> -1.704; movielens: -2.095 -> -2.208 | Mixed |
| Model sparsity and response jointly with HCPF rather than HPF | HCPF versus HPF · test AUC across 12 data sets | AUC | Helped |
Closest papers by meaning
- Coupled Poisson Factorization Integrated With User/Item Metadata for Modeling Popular and Sparse Ratings in Scalable Recommendation (AAAI 2018 · similarity 0.91)
- Gamma-Poisson Dynamic Matrix Factorization Embedded with Metadata Influence (NeurIPS 2018 · similarity 0.90)
- Ordinal Non-negative Matrix Factorization for Recommendation (ICML 2020 · similarity 0.87)
- Hierarchical Negative Binomial Factorization for Recommender Systems on Implicit Feedback (AAAI 2021 · similarity 0.86)
- Recommendations with Sparse Comparison Data: Provably Fast Convergence for Nonconvex Matrix Factorization (ICML 2025 · similarity 0.85)
- Learning-Efficient Yet Generalizable Collaborative Filtering for Item Recommendation (ICML 2024 · 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.