arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.04013cs.CV
收藏

深度神经网络上的隐私泄露:模型反演攻击与防御综述

Privacy Leakage on DNNs: A Survey of Model Inversion Attacks and Defenses

Tsinghua Shenzhen International Graduate School, Tsinghua University(清华大学深圳国际研究生院);School of Computer Science and Technology, Harbin Institute of Technology(哈尔滨工业大学计算机科学与技术学院);Peng Cheng Laboratory(鹏城实验室);Department of Computer Science and Technology, Tsinghua University(清华大学计算机科学与技术系)

Hao Fang, Yixiang Qiu, Hongyao Yu, Wenbo Yu, Jiawei Kong, Baoli Chong, Bin Chen, Xuan Wang, Shu-Tao Xia, Ke Xu

AI总结:

本文系统综述深度神经网络上的模型反演攻击与防御,提出新分类法,并开源工具箱,以应对训练数据隐私泄露威胁。

摘要:

深度神经网络(DNNs)凭借其在众多应用中的卓越性能,彻底改变了各个领域。然而,模型反演(MI)攻击通过滥用对已训练模型的访问权限,泄露关于训练数据集的私有信息,已成为一种严峻的隐私威胁。给定一个训练好的网络,这些攻击使 adversaries 能够重建与私有训练样本高度一致的高保真数据,引发严重的隐私担忧。尽管该领域进展迅速,我们仍缺乏对现有 MI 攻击与防御的全面、系统性的概述。为填补这一空白,本文深入探究该领域,呈现一份整体性综述。首先,我们的工作简要回顾了早期在传统机器学习场景下的 MI 研究。随后,我们详尽分析并比较了近年来针对深度神经网络(DNNs)的众多攻击与防御方法,涵盖多种模态和学习任务。通过细致分析它们的独特特征,我们将这些方法归纳分类为不同类别,并提出一种新颖的分类体系。最后,本文讨论了有前景的研究方向,并为未解决问题提供了潜在解决方案。为促进对 MI 攻击与防御的进一步研究,我们在 GitHub 上实现了一个开源模型反演工具箱(此 https 链接)。

英文摘要:

Deep Neural Networks (DNNs) have revolutionized various domains with their exceptional performance across numerous applications. However, Model Inversion (MI) attacks, which disclose private information about the training dataset by abusing access to the trained models, have emerged as a formidable privacy threat. Given a trained network, these attacks enable adversaries to reconstruct high-fidelity data that closely aligns with the private training samples, posing significant privacy concerns. Despite the rapid advances in the field, we lack a comprehensive and systematic overview of existing MI attacks and defenses. To fill this gap, this paper thoroughly investigates this realm and presents a holistic survey. Firstly, our work briefly reviews early MI studies on traditional machine learning scenarios. We then elaborately analyze and compare numerous recent attacks and defenses on Deep Neural Networks (DNNs) across multiple modalities and learning tasks. By meticulously analyzing their distinctive features, we summarize and classify these methods into different categories and provide a novel taxonomy. Finally, this paper discusses promising research directions and presents potential solutions to open issues. To facilitate further study on MI attacks and defenses, we have implemented an open-source model inversion toolbox on GitHub (https://github.com/ffhibnese/Model-Inversion-Attack-ToolBox).

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。

↑