Research map / Learning theory and algorithms / Clustering
A Model Selection Approach for Clustering a Multinomial Sequence with Non-Negative Factorization
Nam H. Lee, Runze Tang, Carey E. Priebe, Michael Alan Rosen
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 the derived penalty instead of conventional AIC penalty (ablation) | Rank selection by NMF · Simulated d=25 matrix | times K-hat=2 out of 100: 1 -> 61 | Improved |
| use the derived penalty instead of conventional AIC penalty (ablation) | Rank selection by NMF · Simulated d=20 matrix | times K-hat=2 out of 100: 0 -> 11 | Improved |
| use the derived penalty instead of conventional AIC penalty (ablation) | Rank selection by NMF · Simulated d=50 matrix | times K-hat=2 out of 100: 56 -> 100 | Improved |
| use the derived penalty instead of conventional AIC penalty (ablation) | Rank selection by NMF · Simulated d=100 matrix | times K-hat=2 out of 100: 82 -> 100 | Improved |
| use the derived penalty instead of conventional AIC penalty (ablation) | Rank selection by NMF · Single 50x2 matrix | selected K: K=1 -> K=2 | Improved |
| use NMF with the penalized criterion instead of two-sample test (full vertex correspondence) | aic o nmf · Graph clustering n=40 | average ARI: 0.18 -> 0.42 | Improved |
| use NMF with the penalized criterion instead of two-sample test | aic o nmf · Graph clustering n=120 | average ARI: 0.39 -> 0.92 | Improved |
| increase number of vertices n from 40 to 120 | aic o nmf · Graph clustering | average ARI: 0.42 -> 0.92 | Improved |
| known vertex correspondence only for few vertices (seeded matching) | aic o nmf · Graph clustering, small m | average ARI | Hurt |
| use aic o nmf instead of pamk o dist or mclust o pca | aic o nmf · Aggregated Poisson block graphs | ARI | Improved |
Closest papers by meaning
- Fast and Stable Maximum Likelihood Estimation for Incomplete Multinomial Models (ICML 2019 · similarity 0.88)
- Fast Algorithm for Generalized Multinomial Models with Ranking Data (ICML 2019 · similarity 0.86)
- Simultaneous Clustering and Model Selection: Algorithm, Theory and Applications (TPAMI 2017 · similarity 0.86)
- A Unified Framework for Variable Selection in Model-Based Clustering with Missing Not at Random (NeurIPS 2025 · similarity 0.86)
- Simultaneous Clustering and Estimation of Heterogeneous Graphical Models (JMLR 2018 · similarity 0.85)
- Achieving Optimal Clustering in Gaussian Mixture Models with Anisotropic Covariance Structures (NeurIPS 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.
