Lagrangian Inference for Ranking Problems
From MaRDI portal
Publication:6202668
DOI10.1287/OPRE.2022.2313arXiv2110.00151OpenAlexW3203912054MaRDI QIDQ6202668FDOQ6202668
Authors:
Publication date: 26 February 2024
Published in: Operations Research (Search for Journal in Brave)
Abstract: We propose a novel combinatorial inference framework to conduct general uncertainty quantification in ranking problems. We consider the widely adopted Bradley-Terry-Luce (BTL) model, where each item is assigned a positive preference score that determines the Bernoulli distributions of pairwise comparisons' outcomes. Our proposed method aims to infer general ranking properties of the BTL model. The general ranking properties include the "local" properties such as if an item is preferred over another and the "global" properties such as if an item is among the top -ranked items. We further generalize our inferential framework to multiple testing problems where we control the false discovery rate (FDR), and apply the method to infer the top- ranked items. We also derive the information-theoretic lower bound to justify the minimax optimality of the proposed method. We conduct extensive numerical studies using both synthetic and real datasets to back up our theory.
Full work available at URL: https://arxiv.org/abs/2110.00151
Recommendations
- Accelerated MM Algorithms for Inference of Ranking Scores from Comparison Data
- Optimal full ranking from pairwise comparisons
- Simple, Robust and Optimal Ranking from Pairwise Comparisons
- scientific article; zbMATH DE number 5364205
- Spectral method and regularized MLE are both optimal for top-\(K\) ranking
rankingminimax lower boundpairwise comparisonscombinatorial inferenceBradley-Terry-Luce modelmachine learning and data science
Cited In (1)
This page was built for publication: Lagrangian Inference for Ranking Problems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6202668)
