Research map / Learning theory and algorithms / Prediction and decision losses
Efficient Computation of Rankings from Pairwise Comparisons
M. E. J. Newman
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 the new iteration instead of Zermelo's algorithm | Bradley-Terry ranking · Synthetic (MLE without ties) | iterations to convergence: 1270±470 -> 12±2, speed-up ×104 | Improved |
| use the new iteration instead of Zermelo's algorithm | Bradley-Terry ranking · Wolves (MLE without ties) | iterations to convergence: 2410±10 -> 145±1, speed-up ×17 | Improved |
| use the new iteration instead of Zermelo's algorithm | Bradley-Terry ranking · Vervet monkeys (MLE without ties) | iterations to convergence: 232±8 -> 19±1, speed-up ×12 | Improved |
| use the new iteration instead of Zermelo's algorithm | Bradley-Terry ranking · American football (MLE without ties) | iterations to convergence: 49±3 -> 15±1, speed-up ×3.4 | Improved |
| use the new iteration instead of Zermelo's algorithm | Bradley-Terry ranking · Political figures (MLE without ties) | iterations to convergence: 54±3 -> 8±5, speed-up ×7.1 | Improved |
| use the new iteration instead of Zermelo's algorithm | Bradley-Terry ranking · Photographs (MLE without ties) | iterations to convergence: 911±4 -> 22±0, speed-up ×41 | Improved |
| use the new iteration instead of Zermelo's algorithm | Bradley-Terry ranking · Synthetic (MAP without ties) | iterations to convergence: 1560±40 -> 185±18, speed-up ×8.5 | Improved |
| use the new iteration instead of Zermelo's algorithm | Bradley-Terry ranking · Wolves (MAP without ties) | iterations to convergence: 49,200±1700 -> 2200±110, speed-up ×22 | Improved |
| use the new iteration instead of Zermelo's algorithm | Bradley-Terry ranking · American football (MAP without ties) | iterations to convergence: 19,200±1500 -> 6000±600, speed-up ×3.3 | Improved |
| use the new iteration instead of Zermelo's algorithm | Bradley-Terry ranking · Photographs (MAP without ties) | iterations to convergence: 1186±3 -> 82±10, speed-up ×14 | Improved |
| use the new iteration instead of Zermelo's algorithm | Bradley-Terry ranking · Synthetic (with ties) | iterations to convergence: 1130±760 -> 27±8, speed-up ×42 | Improved |
| use the new iteration instead of Zermelo's algorithm | Bradley-Terry ranking · Soccer (with ties) | iterations to convergence: 1650±16 -> 421±5, speed-up ×3.9 | Improved |
Closest papers by meaning
- Simple, Robust and Optimal Ranking from Pairwise Comparisons (JMLR 2018 · similarity 0.91)
- On Sample Complexity Upper and Lower Bounds for Exact Ranking from Noisy Comparisons (NeurIPS 2019 · similarity 0.89)
- Generalized Results for the Existence and Consistency of the MLE in the Bradley-Terry-Luce Model (ICML 2022 · similarity 0.89)
- Dynamic Ranking with the BTL Model: A Nearest Neighbor based Rank Centrality Method (JMLR 2023 · similarity 0.88)
- Foundations of the Theory of Performance-Based Ranking (CVPR 2025 · similarity 0.86)
- Active Ranking and Matchmaking, with Perfect Matchings (ICML 2024 · similarity 0.85)
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.
