Research map / Learning theory and algorithms / Prediction and decision losses
Data-driven Rank Breaking for Efficient Rank Aggregation
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 |
|---|---|---|---|
| Use data-driven weights instead of uniform weights for consistent rank-breaking (ablation) | Rank-breaking on top-ℓ separators · Top-ℓ separators, simulated PL data | mean squared error | Helped |
| Use data-driven rank-breaking instead of GMM | PL rank aggregation · Sushi preferences | mean squared error | Helped |
| Use data-driven rank-breaking instead of GMM | PL rank aggregation · Jester | mean squared error | Helped |
| Use data-driven rank-breaking instead of GMM | PL rank aggregation · APA election preferences | mean squared error | Helped |
| Use data-driven rank-breaking instead of GMM | PL rank aggregation · Sushi preferences | Kendall rank correlation | Helped |
| Use data-driven rank-breaking instead of other aggregation methods | Sushi preferences · Sushi, n = 5000, ℓ = 5 | Kendall rank correlation: 0.217 vs 0.200 | Mixed |
| Choose separator positions uniformly across the offered items instead of restricting them to the top (ablation) | Data-driven rank-breaking · Simulated PL rankings | mean squared error | Helped |
| Place separators at the bottom rather than across the ranking (ablation) | Data-driven rank-breaking · Bottom-ℓ separators | mean squared error | Hurt |
| Use a comparison graph with a larger spectral gap (ablation) | Data-driven rank-breaking · Simulated comparison graph topologies | mean squared error | Helped |
| Use random pairwise comparisons with uniform weights instead of consistent data-driven rank-breaking (ablation) | Rank-breaking · Random separator positions | mean squared error | Mixed |
| Use optimal weights instead of uniform weights when offering-set sizes are heterogeneous (ablation) | Consistent rank-breaking · Mixed top-ℓ and top-1 comparisons | mean squared error: roughly 10 times worse | Helped |
| Use optimal weights instead of uniform weights with fixed offering-set size (ablation) | Consistent rank-breaking · Separators at the top and bottom | mean squared error | Helped |
Closest papers by meaning
- Accelerated Spectral Ranking (ICML 2018 · similarity 0.89)
- Computational and Statistical Tradeoffs in Learning to Rank (NeurIPS 2016 · similarity 0.89)
- Selective Preference Aggregation (ICML 2025 · similarity 0.87)
- Aggregating Quantitative Relative Judgments: From Social Choice to Ranking Prediction (NeurIPS 2024 · similarity 0.86)
- Identifiability Matters: Revealing the Hidden Recoverable Condition in Unbiased Learning to Rank (ICML 2024 · similarity 0.86)
- Stability and Multigroup Fairness in Ranking with Uncertain Predictions (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.