arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2205.06369cs.LGcs.CR
收藏

如何在多个更新模型上组合成员推断攻击

How to Combine Membership-Inference Attacks on Multiple Updated Models

Google, Inc.(谷歌公司);Khoury College of Computer Sciences, Northeastern University(东北大学库里计算机学院);Columbia University(哥伦比亚大学)

Matthew Jagielski, Stanley Wu, Alina Oprea, Jonathan Ullman, Roxana Geambasu

摘要:

大量研究表明,机器学习模型容易受到成员推断(MI)攻击,这些攻击侵犯了训练数据中参与者的隐私。大多数MI研究都集中在单个独立模型的情况下,而生产机器学习平台通常会随着时间的推移更新模型,数据通常会在分布中发生变化,从而为攻击者提供更多信息。本文提出了新的攻击,利用一个或多个模型更新,以提高MI。我们的方法的一个关键部分是利用丰富的信息,从独立的MI攻击单独安装对原始和更新的模型,并结合这些信息以特定的方式来提高攻击的有效性。我们提出了一套组合功能和调整方法,并提出了各种选择的分析和定量的理由。我们在四个公共数据集上的结果表明,我们的攻击在使用更新信息方面是有效的,与对独立模型的攻击相比,我们的攻击具有显着的优势,但也与利用相关机器学习设置中的模型更新的先前MI攻击相比。我们进行了第一次测量的影响,分布偏移MI攻击模型更新,并表明,一个更剧烈的分布偏移的结果在显着更高的MI风险比逐渐转变。我们的代码可在https://www.github.com/stanleykywu/model-updates上获得。

英文摘要:

A large body of research has shown that machine learning models are vulnerable to membership inference (MI) attacks that violate the privacy of the participants in the training data. Most MI research focuses on the case of a single standalone model, while production machine-learning platforms often update models over time, on data that often shifts in distribution, giving the attacker more information. This paper proposes new attacks that take advantage of one or more model updates to improve MI. A key part of our approach is to leverage rich information from standalone MI attacks mounted separately against the original and updated models, and to combine this information in specific ways to improve attack effectiveness. We propose a set of combination functions and tuning methods for each, and present both analytical and quantitative justification for various options. Our results on four public datasets show that our attacks are effective at using update information to give the adversary a significant advantage over attacks on standalone models, but also compared to a prior MI attack that takes advantage of model updates in a related machine-unlearning setting. We perform the first measurements of the impact of distribution shift on MI attacks with model updates, and show that a more drastic distribution shift results in significantly higher MI risk than a gradual shift. Our code is available at https://www.github.com/stanleykywu/model-updates.

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。

↑