Research map / Learning theory and algorithms / Matrix and tensor methods
Fast and Stable Maximum Likelihood Estimation for Incomplete Multinomial Models
Chenyang Zhang, Guosheng Yin
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 stable weaver instead of Bayesian weaver | incomplete multinomial MLE · Gender-age simulation | iterations: 439.23 -> 20.32 | Improved |
| use stable weaver instead of self-consistency | incomplete multinomial MLE · Gender-age simulation | iterations: 47.65 -> 20.32 | Improved |
| use stable weaver instead of basic weaver | incomplete multinomial MLE · Gender-age simulation | iterations: 16.25 -> 20.32 | Hurt |
| use stable weaver instead of Bayesian weaver and self-consistency | K=20000 weak signal · Weak signal simulation | iterations to convergence: 2229 -> 45 | Improved |
| use stable weaver instead of self-consistency | K=20000 weak signal · Weak signal simulation | iterations to convergence: 3438 -> 45 | Improved |
| use stable weaver instead of Bayesian weaver | NASCAR (w/o ties) · Table 5 | time (s): 25.27 -> <0.01 | Improved |
| use stable weaver instead of MM | HKJC1416 (w/o ties) · Table 5 | time (s): 375.79 -> 38.46 | Improved |
| use stable weaver instead of ILSR | HKJC1416 (w/o ties) · Table 5 | time (s): 1166.97 -> 38.46 | Improved |
| use stable weaver instead of trust region | HKJC1416 (w/ ties) · Table 5 | time (s): 1835.37 -> 86.40 | Improved |
| use ILSR instead of stable weaver in iteration count | NASCAR (w/o ties) · Table 5 | iterations: 22 -> 12 | Hurt |
| increase R from 16 to 128 with m=1 (ablation) | stable weaver · Random partition K=32 | variance of estimated p1 | Improved |
| increase m from 1 to 50 with R=16 (ablation) | stable weaver · Random partition K=32 | variance of estimated p1 | Improved |
Closest papers by meaning
- Fast Algorithm for Generalized Multinomial Models with Ranking Data (ICML 2019 · similarity 0.90)
- A Model Selection Approach for Clustering a Multinomial Sequence with Non-Negative Factorization (TPAMI 2016 · similarity 0.88)
- Statistical Inference for Incomplete Ranking Data: The Case of Rank-Dependent Coarsening (ICML 2017 · similarity 0.86)
- Generalized Rank-Breaking: Computational and Statistical Tradeoffs (JMLR 2018 · similarity 0.85)
- Score Matching with Missing Data (ICML 2025 · similarity 0.84)
- A Pairwise Pseudo-likelihood Approach for Matrix Completion with Informative Missingness (NeurIPS 2024 · similarity 0.82)
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.
