Research map / Learning theory and algorithms / Prediction and decision losses
Active Learning for Top-$K$ Rank Aggregation from Noisy Comparisons
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 |
|---|---|---|---|
| Adaptive (active) pairwise comparisons instead of passive measurements | top-K ranking under uniform noise model · top-K sorting, uniform noise | sample complexity gain: Omega(n^2 log n / log log n) factor gain | Helped |
| Adaptive (active) pairwise comparisons instead of passive measurements | top-K ranking under BTL model · top-K sorting, BTL model | sample complexity gain: Omega(log n / log log n) factor gain | Helped |
| Repeating each pairwise comparison m times and using majority vote in the single-elimination tournament and heap | proposed top-K algorithm · top-K sorting and partitioning, noisy comparisons | sample complexity scaling | Helped |
| Proposed top-K algorithm vs Braverman et al. extended to top-K, for K=16 under uniform noise | top-K identification · n=1024 uniform noise, top-16 | total comparisons to reach target success rate | Helped |
| Proposed top-K algorithm vs Heckel et al. algorithm | top-K identification · uniform noise and BTL model, n=1024 | success rate vs total comparisons | Helped |
| Top-1 proposed algorithm vs Braverman et al. (K=1) | top-1 selection under uniform noise · n=1024 uniform noise, top-1 | total comparisons to reach target success rate | No clear effect |
| Partitioning vs sorting in proposed algorithm (ablation) | top-K algorithm · uniform noise, n=1024, top-16 | success rate vs comparisons | No clear effect |
Closest papers by meaning
- Active Ranking without Strong Stochastic Transitivity (NeurIPS 2022 · similarity 0.90)
- Sample Complexity Bounds for Active Ranking from Multi-wise Comparisons (NeurIPS 2021 · similarity 0.88)
- Just Sort It! A Simple and Effective Approach to Active Preference Learning (ICML 2017 · similarity 0.87)
- Optimal Top-$k$ Identification from Pairwise Comparisons (ICML 2026 · similarity 0.87)
- Active Ranking and Matchmaking, with Perfect Matchings (ICML 2024 · similarity 0.86)
- Ranking with Multiple Oracles: From Weak to Strong Stochastic Transitivity (ICML 2025 · similarity 0.84)
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.