Generalized Priority-Aware Shapley Value

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Lee, Kiljae, Liu, Ziqi, Tang, Weijing, Zhang, Yuan
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866917496123031552
author Lee, Kiljae
Liu, Ziqi
Tang, Weijing
Zhang, Yuan
author_facet Lee, Kiljae
Liu, Ziqi
Tang, Weijing
Zhang, Yuan
contents Shapley value and its priority-aware extensions are widely used for valuation in machine learning, but existing methods require pairwise priority to be binary and acyclic, a restriction spectacularly violated in real-data examples such as aggregated human preferences and multi-criterion comparisons. We introduce the generalized priority-aware Shapley value (GPASV), a random order value defined on arbitrary directed weighted priority graphs, in which pairwise edges penalize rather than forbid order violations. GPASV covers a range of classical models as boundary cases. We establish GPASV through an axiomatic characterization, develop the associated computational methods, and introduce a priority sweeping diagnostic extending PASV's. We apply GPASV to LLM ensemble valuation on the cyclic Chatbot Arena preference graph, illustrating that priority-aware valuation is not a one-button operation: different balances of pairwise graph priority versus individual soft priority produce substantively different valuations of the same data.
format Preprint
id arxiv_https___arxiv_org_abs_2605_15018
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Generalized Priority-Aware Shapley Value
Lee, Kiljae
Liu, Ziqi
Tang, Weijing
Zhang, Yuan
Machine Learning
Artificial Intelligence
Shapley value and its priority-aware extensions are widely used for valuation in machine learning, but existing methods require pairwise priority to be binary and acyclic, a restriction spectacularly violated in real-data examples such as aggregated human preferences and multi-criterion comparisons. We introduce the generalized priority-aware Shapley value (GPASV), a random order value defined on arbitrary directed weighted priority graphs, in which pairwise edges penalize rather than forbid order violations. GPASV covers a range of classical models as boundary cases. We establish GPASV through an axiomatic characterization, develop the associated computational methods, and introduce a priority sweeping diagnostic extending PASV's. We apply GPASV to LLM ensemble valuation on the cyclic Chatbot Arena preference graph, illustrating that priority-aware valuation is not a one-button operation: different balances of pairwise graph priority versus individual soft priority produce substantively different valuations of the same data.
title Generalized Priority-Aware Shapley Value
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2605.15018