Bounded-Abstention Pairwise Learning to Rank

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Ferrara, Antonio, Pugnana, Andrea, Bonchi, Francesco, Ruggieri, Salvatore
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913866034708480
author Ferrara, Antonio
Pugnana, Andrea
Bonchi, Francesco
Ruggieri, Salvatore
author_facet Ferrara, Antonio
Pugnana, Andrea
Bonchi, Francesco
Ruggieri, Salvatore
contents Ranking systems influence decision-making in high-stakes domains like health, education, and employment, where they can have substantial economic and social impacts. This makes the integration of safety mechanisms essential. One such mechanism is $\textit{abstention}$, which enables algorithmic decision-making system to defer uncertain or low-confidence decisions to human experts. While abstention have been predominantly explored in the context of classification tasks, its application to other machine learning paradigms remains underexplored. In this paper, we introduce a novel method for abstention in pairwise learning-to-rank tasks. Our approach is based on thresholding the ranker's conditional risk: the system abstains from making a decision when the estimated risk exceeds a predefined threshold. Our contributions are threefold: a theoretical characterization of the optimal abstention strategy, a model-agnostic, plug-in algorithm for constructing abstaining ranking models, and a comprehensive empirical evaluations across multiple datasets, demonstrating the effectiveness of our approach.
format Preprint
id arxiv_https___arxiv_org_abs_2505_23437
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Bounded-Abstention Pairwise Learning to Rank
Ferrara, Antonio
Pugnana, Andrea
Bonchi, Francesco
Ruggieri, Salvatore
Machine Learning
Artificial Intelligence
Information Retrieval
Ranking systems influence decision-making in high-stakes domains like health, education, and employment, where they can have substantial economic and social impacts. This makes the integration of safety mechanisms essential. One such mechanism is $\textit{abstention}$, which enables algorithmic decision-making system to defer uncertain or low-confidence decisions to human experts. While abstention have been predominantly explored in the context of classification tasks, its application to other machine learning paradigms remains underexplored. In this paper, we introduce a novel method for abstention in pairwise learning-to-rank tasks. Our approach is based on thresholding the ranker's conditional risk: the system abstains from making a decision when the estimated risk exceeds a predefined threshold. Our contributions are threefold: a theoretical characterization of the optimal abstention strategy, a model-agnostic, plug-in algorithm for constructing abstaining ranking models, and a comprehensive empirical evaluations across multiple datasets, demonstrating the effectiveness of our approach.
title Bounded-Abstention Pairwise Learning to Rank
topic Machine Learning
Artificial Intelligence
Information Retrieval
url https://arxiv.org/abs/2505.23437