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 2104.05996cs.CRcs.LG
收藏

巨人的陨落:流行的基于文本的MLaaS如何抵御简单的规避攻击

Fall of Giants: How popular text-based MLaaS fall against a simple evasion attack

University of Padua(帕多瓦大学)

Luca Pajola, Mauro Conti

摘要:

对机器学习应用的需求增加使得公司提供机器学习即服务(MLaaS)。在MLaaS(到2025年市场估计为8亿美元)中,用户为性能良好的ML模型付费,而无需处理复杂的训练过程。在MLaaS中,基于文本的应用程序是最受欢迎的应用程序(例如,语言翻译)。鉴于这种流行,MLaaS必须为对抗性操作提供弹性。例如,错误的翻译可能会导致双方之间的误解。在文本领域,最先进的攻击主要集中在利用ML模型弱点的策略上。不幸的是,没有太多的注意力给予其他流水线的阶段,如索引阶段(即,当句子从文本表示转换为数字表示时),如果操纵,则会显著影响应用程序的最终性能。在本文中,我们提出了一种新的文本规避技术称为“\textit{Zero-Width}攻击”(ZeW),利用人类不可读字符的注入,影响索引阶段的机制。我们证明了我们简单而有效的攻击欺骗了亚马逊,谷歌,IBM和微软等“巨头”的MLaaS。我们的案例研究基于对仇恨推文的操纵,表明在12个分析的服务中,只有一个对我们的注入策略有抵抗力。最后,我们介绍和测试一个简单的\textit{input validation}防御,可以防止我们提出的攻击。

英文摘要:

The increased demand for machine learning applications made companies offer Machine-Learning-as-a-Service (MLaaS). In MLaaS (a market estimated 8000M USD by 2025), users pay for well-performing ML models without dealing with the complicated training procedure. Among MLaaS, text-based applications are the most popular ones (e.g., language translators). Given this popularity, MLaaS must provide resiliency to adversarial manipulations. For example, a wrong translation might lead to a misunderstanding between two parties. In the text domain, state-of-the-art attacks mainly focus on strategies that leverage ML models' weaknesses. Unfortunately, not much attention has been given to the other pipeline' stages, such as the indexing stage (i.e., when a sentence is converted from a textual to a numerical representation) that, if manipulated, can significantly affect the final performance of the application. In this paper, we propose a novel text evasion technique called "\textit{Zero-Width} attack" (ZeW) that leverages the injection of human non-readable characters, affecting indexing stage mechanisms. We demonstrate that our simple yet effective attack deceives MLaaS of "giants" such as Amazon, Google, IBM, and Microsoft. Our case study, based on the manipulation of hateful tweets, shows that out of 12 analyzed services, only one is resistant to our injection strategy. We finally introduce and test a simple \textit{input validation} defense that can prevent our proposed attack.

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。

↑