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 2107.01614cs.LG
收藏

调查:推理时的泄漏和隐私

Survey: Leakage and Privacy at Inference Time

Facebook FAIAR(脸书FAIAR);University of Edinburgh(爱丁堡大学);University of Glasgow(格拉斯哥大学);NHS Scotland(苏格兰国民保健制度);Canon Medical Research Europe(佳能医疗欧洲研究中心);The Alan Turing Institute(艾伦·图灵研究所)

Marija Jegorova, Chaitanya Kaul, Charlie Mayor, Alison Q. O'Neil, Alexander Weir, Roderick Murray-Smith, Sotirios A. Tsaftaris

摘要:

从公开的机器学习(ML)模型中泄漏数据是一个日益重要的领域,因为ML的商业和政府应用可以利用多个数据源,可能包括用户和客户的敏感数据。我们对几个方面的当代进展进行了全面的调查,包括ML模型自然发生的非自愿数据泄漏,隐私攻击造成的潜在恶意泄漏以及目前可用的防御机制。我们专注于推理时间泄漏,作为最有可能的情况下公开可用的模型。我们首先讨论在不同的数据、任务和模型架构的上下文中泄漏是什么。然后,我们提出了一个分类在非自愿和恶意泄漏,可用的防御,其次是目前可用的评估指标和应用程序。我们总结了突出的挑战和开放的问题,概述了未来研究的一些有前途的方向。

英文摘要:

Leakage of data from publicly available Machine Learning (ML) models is an area of growing significance as commercial and government applications of ML can draw on multiple sources of data, potentially including users' and clients' sensitive data. We provide a comprehensive survey of contemporary advances on several fronts, covering involuntary data leakage which is natural to ML models, potential malevolent leakage which is caused by privacy attacks, and currently available defence mechanisms. We focus on inference-time leakage, as the most likely scenario for publicly available models. We first discuss what leakage is in the context of different data, tasks, and model architectures. We then propose a taxonomy across involuntary and malevolent leakage, available defences, followed by the currently available assessment metrics and applications. We conclude with outstanding challenges and open questions, outlining some promising directions for future research.

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。

↑