Research map / Learning theory and algorithms / Prediction and decision losses
Mallows ranking models: maximum likelihood estimate and regeneration
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 |
|---|---|---|---|
| select model size automatically rather than fix t=1 | IGM ranking model · Synthetic IGM, t_max=20 and t_max=40 | estimated ranking accuracy: 94 % -> 100 %; 82 % -> 100 % | Helped |
| select model size automatically rather than fix t=10 | IGM ranking model · Synthetic IGM, t_max=10, t_max=20, and t_max=40 | estimated ranking accuracy: 100 % at all three settings | No clear effect |
| select model size automatically rather than fix t=10 | IGM ranking model · Synthetic IGM, t_max=10, t_max=20, and t_max=40 | average training time: 14.45 s -> 2.80 s; 54.45 s -> 24.42 s; 684.65 s -> 391.20 s | Helped |
| select model size automatically rather than fix t=1 | IGM ranking model · Synthetic IGM, t_max=10 | average training time: 1.56 s -> 2.80 s | Hurt |
| select model size automatically rather than fix t=1 | IGM ranking model · Synthetic IGM, t_max=20 and t_max=40 | average training time: 5.73 s -> 24.42 s; 70.26 s -> 391.20 s | Hurt |
| select model size automatically rather than fix t=30 | IGM ranking model · University homepage search | mean target rank: 18.9 -> 6.8 | Helped |
| select model size automatically rather than fix t=10 | IGM ranking model · University homepage search | mean target rank: 13.2 -> 6.8 | Helped |
| select model size automatically rather than fix t=1 | IGM ranking model · University homepage search | mean target rank: 3.7 -> 6.8 | Hurt |
| select model size automatically rather than fix t=2 | IGM ranking model · University homepage search | median target rank: 2 -> 3 | Hurt |
Closest papers by meaning
- Concentric mixtures of Mallows models for top-$k$ rankings: sampling and identifiability (ICML 2021 · similarity 0.88)
- Properties of the Mallows Model Depending on the Number of Alternatives: A Warning for an Experimentalist (ICML 2023 · similarity 0.88)
- Ranking Distributions based on Noisy Sorting (ICML 2018 · similarity 0.88)
- Mallows Models for Top-k Lists (NeurIPS 2018 · similarity 0.87)
- Generalized Top-k Mallows Model for Ranked Choices (NeurIPS 2025 · similarity 0.86)
- Beyond Kemeny Medians: Consensus Ranking Distributions. Definition, Properties and Statistical Learning (NeurIPS 2026 · 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.