Shapley-Guided Utility Learning for Effective Graph Inference Data Valuation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chi, Hongliang, Wu, Qiong, Zhou, Zhengyi, Ma, Yao
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913754002751488
author Chi, Hongliang
Wu, Qiong
Zhou, Zhengyi
Ma, Yao
author_facet Chi, Hongliang
Wu, Qiong
Zhou, Zhengyi
Ma, Yao
contents Graph Neural Networks (GNNs) have demonstrated remarkable performance in various graph-based machine learning tasks, yet evaluating the importance of neighbors of testing nodes remains largely unexplored due to the challenge of assessing data importance without test labels. To address this gap, we propose Shapley-Guided Utility Learning (SGUL), a novel framework for graph inference data valuation. SGUL innovatively combines transferable data-specific and modelspecific features to approximate test accuracy without relying on ground truth labels. By incorporating Shapley values as a preprocessing step and using feature Shapley values as input, our method enables direct optimization of Shapley value prediction while reducing computational demands. SGUL overcomes key limitations of existing methods, including poor generalization to unseen test-time structures and indirect optimization. Experiments on diverse graph datasets demonstrate that SGUL consistently outperforms existing baselines in both inductive and transductive settings. SGUL offers an effective, efficient, and interpretable approach for quantifying the value of test-time neighbors.
format Preprint
id arxiv_https___arxiv_org_abs_2503_18195
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Shapley-Guided Utility Learning for Effective Graph Inference Data Valuation
Chi, Hongliang
Wu, Qiong
Zhou, Zhengyi
Ma, Yao
Machine Learning
Computer Science and Game Theory
Graph Neural Networks (GNNs) have demonstrated remarkable performance in various graph-based machine learning tasks, yet evaluating the importance of neighbors of testing nodes remains largely unexplored due to the challenge of assessing data importance without test labels. To address this gap, we propose Shapley-Guided Utility Learning (SGUL), a novel framework for graph inference data valuation. SGUL innovatively combines transferable data-specific and modelspecific features to approximate test accuracy without relying on ground truth labels. By incorporating Shapley values as a preprocessing step and using feature Shapley values as input, our method enables direct optimization of Shapley value prediction while reducing computational demands. SGUL overcomes key limitations of existing methods, including poor generalization to unseen test-time structures and indirect optimization. Experiments on diverse graph datasets demonstrate that SGUL consistently outperforms existing baselines in both inductive and transductive settings. SGUL offers an effective, efficient, and interpretable approach for quantifying the value of test-time neighbors.
title Shapley-Guided Utility Learning for Effective Graph Inference Data Valuation
topic Machine Learning
Computer Science and Game Theory
url https://arxiv.org/abs/2503.18195