Research map / Learning theory and algorithms / Prediction and decision losses
Just Sort It! A Simple and Effective Approach to Active Preference Learning
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 sorting-based adaptive pair selection instead of random sampling | Quicksort and Mergesort active learning · GIFGIF | displacement: 14~\% larger for random sampling than Quicksort; 23~\% larger than Mergesort | Helped |
| Use Mergesort instead of Quicksort to select comparisons (ablation) | Sorting-based active learning · Synthetic dataset ($n = 200$, $\lambda = 5$) | displacement | Helped |
| Use sorting-based active learning instead of random sampling | Preference ranking · Synthetic dataset ($n = 200$, $\lambda = 5$) | displacement | Helped |
| Use sorting-based active learning instead of Bayesian active learning | Preference ranking · Synthetic dataset ($n = 200$, $\lambda = 5$) | displacement | Hurt |
| Use sorting-based active learning instead of random sampling | Preference ranking · Sushi | displacement | Helped |
| Use sorting-based active learning instead of Bayesian active learning | Preference ranking · Sushi | displacement | No clear effect |
| Use uncertainty sampling as the comparison-selection strategy | Preference ranking · Sushi | displacement | Hurt |
| Use sorting-based adaptive pair selection instead of random sampling | Preference ranking · GIFGIF | displacement: 14~\% larger for random sampling than Quicksort; 23~\% larger than Mergesort | Helped |
| Use adjacent-neighbor uncertainty sampling (ablation) | Uncertainty sampling · GIFGIF | displacement: over $5\times$ larger than random sampling for $c = 10^6$ | Hurt |
| Use Mergesort instead of Quicksort to select comparisons (ablation) | Sorting-based active learning · GIFGIF | comparisons to reach a target displacement: approximately $2 \times$ fewer comparisons than random sampling | Helped |
Closest papers by meaning
- Active preference learning for ordering items in- and out-of-sample (NeurIPS 2024 · similarity 0.91)
- Preference Elicitation as Average-Case Sorting (AAAI 2021 · similarity 0.89)
- TopRank: A practical algorithm for online stochastic ranking (NeurIPS 2018 · similarity 0.88)
- Comparing Few to Rank Many: Active Human Preference Learning Using Randomized Frank-Wolfe Method (ICML 2025 · similarity 0.86)
- Active Ranking and Matchmaking, with Perfect Matchings (ICML 2024 · similarity 0.85)
- Preference Learning Algorithms Do Not Learn Preference Rankings (NeurIPS 2024 · 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.