Research map / Learning theory and algorithms / Sample complexity
Two-Sample Testing on Ranked Preference Data and the Role of Modeling Assumptions
Charvi Rastogi, Sivaraman Balakrishnan, Nihar B. Shah, Aarti Singh
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 |
|---|---|---|---|
| scale sample size as n = 1/(a d eps^2) while varying d, eps or a (ablation) | Permutation two-sample test · Synthetic pairwise comparisons | power: curves across all values of varied parameters nearly coincide | Improved |
| generate P and Q under BTL or SST instead of model-free (ablation) | Permutation two-sample test · Synthetic, d=20, eps^2=0.05, a=1 | power: power identical in all the models considered | No clear effect |
| use Complete rank breaking instead of Random Disjoint (ablation) | Ranking two-sample test · Sushi preference, d=10 | empirical power: Complete performs competitively with Kendall and Mallows kernel tests | Improved |
| apply test to ordinal vs cardinal-converted-to-ordinal data pooled over six experiments | Permutation two-sample test · Crowdsourcing ordinal vs cardinal, d=74 | p-value: p-value of 0.003, rejects null | Improved |
| apply test to football league data across consecutive seasons | Permutation two-sample test · European football leagues, d=67 | p-value: p-value of 0.971, fails to reject | No clear effect |
Closest papers by meaning
- Estimation from Pairwise Comparisons: Sharp Minimax Bounds with Topology Dependence (JMLR 2016 · similarity 0.89)
- Sample Complexity Bounds for Active Ranking from Multi-wise Comparisons (NeurIPS 2021 · similarity 0.87)
- Optimal Sample Complexity of M-wise Data for Top-K Ranking (NeurIPS 2017 · similarity 0.86)
- Generalized Results for the Existence and Consistency of the MLE in the Bradley-Terry-Luce Model (ICML 2022 · similarity 0.86)
- On Sample Complexity Upper and Lower Bounds for Exact Ranking from Noisy Comparisons (NeurIPS 2019 · similarity 0.86)
- Aggregating Quantitative Relative Judgments: From Social Choice to Ranking Prediction (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.
