arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.08679cs.LGcs.AIcs.CRcs.CYstat.ML
收藏

机器学习中的隐私问题综述

An Overview of Privacy in Machine Learning

UCL(伦敦大学学院);Alan Turing Institute(艾伦·图灵研究所)

Emiliano De Cristofaro

摘要:

在过去的几年里,谷歌、微软和亚马逊等供应商已经开始为客户提供软件接口,使他们能够轻松地将机器学习任务嵌入到他们的应用程序中。总体而言,组织现在可以使用机器学习即服务(MLaaS)引擎来外包复杂的任务,例如,训练分类器,执行预测,聚类等,它们还可以让其他人查询在其数据上训练的模型。当然,这种方法也可以在其他情况下使用(并且经常被提倡),包括政府合作,公民科学项目和企业对企业的伙伴关系。然而,如果恶意用户能够恢复用于训练这些模型的数据,则由此产生的信息泄漏将造成严重的问题。同样,如果模型的内部参数被认为是专有信息,那么对模型的访问不应该允许对手学习这些参数。在本文中,我们将回顾在这个空间中的隐私挑战,提供了一个系统的回顾相关的研究文献,并探讨可能的对策。更具体地说,我们提供了有关机器学习和隐私相关概念的丰富背景信息。然后,我们讨论了可能的对抗模型和设置,涵盖了广泛的攻击,涉及到私人和/或敏感信息泄漏,并审查最近的结果,试图抵御这种攻击。最后,我们总结了一系列需要更多工作的开放问题,包括需要更好的评估,更有针对性的防御,以及与政策和数据保护工作的关系的研究。

英文摘要:

Over the past few years, providers such as Google, Microsoft, and Amazon have started to provide customers with access to software interfaces allowing them to easily embed machine learning tasks into their applications. Overall, organizations can now use Machine Learning as a Service (MLaaS) engines to outsource complex tasks, e.g., training classifiers, performing predictions, clustering, etc. They can also let others query models trained on their data. Naturally, this approach can also be used (and is often advocated) in other contexts, including government collaborations, citizen science projects, and business-to-business partnerships. However, if malicious users were able to recover data used to train these models, the resulting information leakage would create serious issues. Likewise, if the inner parameters of the model are considered proprietary information, then access to the model should not allow an adversary to learn such parameters. In this document, we set to review privacy challenges in this space, providing a systematic review of the relevant research literature, also exploring possible countermeasures. More specifically, we provide ample background information on relevant concepts around machine learning and privacy. Then, we discuss possible adversarial models and settings, cover a wide range of attacks that relate to private and/or sensitive information leakage, and review recent results attempting to defend against such attacks. Finally, we conclude with a list of open problems that require more work, including the need for better evaluations, more targeted defenses, and the study of the relation to policy and data protection efforts.

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。

↑