arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.00407cs.LG
收藏

Fed-Listing:图神经网络中的联邦标签分布推断

Fed-Listing: Federated Label Distribution Inference in Graph Neural Networks

Department of Computer Science, University of Nevada Las Vegas(内华达大学拉斯维加斯分校计算机科学系);Department of Computer Science, University of Southern California(南加州大学计算机科学系)

Suprim Nakarmi, Junggab Son, Yue Zhao, Zuobin Xiong

AI总结:

本文提出Fed-Listing,一种基于梯度的攻击方法,用于在联邦图神经网络中推断目标客户端的隐私标签统计信息,展示了其在非独立同分布场景下的优越性能。

摘要:

联邦图神经网络(FedGNNs)在保护用户隐私的前提下,促进多个客户端在图结构数据上的协同学习。然而,研究发现共享模型更新,尤其是梯度,可能无意中泄露本地用户的敏感信息。本文提出Fed-Listing,一种新的基于梯度的攻击方法,旨在在不访问原始数据或节点特征的情况下推断FedGNNs中目标客户端的隐私标签统计信息。Fed-Listing仅利用训练过程中交换的最终层梯度,以隐蔽的方式揭示类别比例的统计模式。在四个基准数据集和三种GNN架构上的广泛实验表明,Fed-Listing在非独立同分布场景下显著优于现有基线,包括随机猜测和Decaf。此外,现有防御机制难以降低Fed-Listing的攻击性能,除非模型的实用性严重受损。代码实现和补充材料可在https://github.com/suprimnakarmi/Fed-Listing获取。

英文摘要:

Federated Graph Neural Networks (FedGNNs) facilitate collaborative learning across multiple clients with graph-structured data while preserving user privacy. However, emerging research indicates that within this setting, shared model updates, particularly gradients, can unintentionally leak sensitive information of local users. Numerous privacy inference attacks have been explored in traditional federated learning and extended to graph settings, but the problem of label distribution inference in FedGNNs remains largely underexplored. In this work, we introduce Fed-Listing (Federated Label Distribution Inference in GNNs), a novel gradient-based attack designed to infer the private label statistics of target clients in FedGNNs without access to raw data or node features. Fed-Listing only leverages the final-layer gradients exchanged during training to uncover statistical patterns that reveal class proportions in a stealthy manner. Extensive experiments on four benchmark datasets and three GNN architectures show that Fed-Listing significantly outperforms existing baselines, including random guessing and Decaf, even under challenging non-i.i.d. scenarios. Moreover, existing defense mechanisms can barely reduce the attack performance of Fed-Listing, unless the model's utility is severely degraded. The code implementation and Supplementary materials are available here: https://github.com/suprimnakarmi/Fed-Listing.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑