arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2009.13243cs.CR
收藏

利用可解释性为恶意软件分类器生成端到端对抗性示例

Generating End-to-End Adversarial Examples for Malware Classifiers Using Explainability

Ishai Rosenberg, Shai Meir, Jonathan Berrebi, Ilay Gordon, Guillaume Sicard, Eli David

摘要:

近年来,可解释机器学习(ML)的研究得到了广泛的关注。到目前为止,这项研究主要集中在常规的ML用户用例上,例如调试ML模型。本文采取了不同的姿态,并表明对手可以利用可解释的ML来绕过多特征类型的恶意软件分类器。先前针对此类分类器的对抗攻击仅添加新特征而不修改现有特征以避免损害经修改的恶意软件可执行文件的功能。当前的攻击使用单一算法,该算法既选择要修改的特征,又盲目地修改它们,对所有特征都一视同仁。在本文中,我们提出了一种不同的方法。我们将对抗性示例生成任务分为两部分:首先,我们使用可解释性算法找到特定样本的所有特征的重要性,然后我们逐个特征地进行特定特征的修改。为了在黑盒场景中应用我们的攻击,我们引入了可解释性的可转移性的概念,即,将可解释性算法应用于使用不同特征子集并在不同数据集上训练的不同分类器,仍然会产生相似的重要特征子集。我们的结论是,可解释性算法可以被对手利用,因此训练更多可解释性分类器的倡导者应该考虑这些分类器对对抗攻击的更高脆弱性的权衡。

英文摘要:

In recent years, the topic of explainable machine learning (ML) has been extensively researched. Up until now, this research focused on regular ML users use-cases such as debugging a ML model. This paper takes a different posture and show that adversaries can leverage explainable ML to bypass multi-feature types malware classifiers. Previous adversarial attacks against such classifiers only add new features and not modify existing ones to avoid harming the modified malware executable's functionality. Current attacks use a single algorithm that both selects which features to modify and modifies them blindly, treating all features the same. In this paper, we present a different approach. We split the adversarial example generation task into two parts: First we find the importance of all features for a specific sample using explainability algorithms, and then we conduct a feature-specific modification, feature-by-feature. In order to apply our attack in black-box scenarios, we introduce the concept of transferability of explainability, that is, applying explainability algorithms to different classifiers using different features subsets and trained on different datasets still result in a similar subset of important features. We conclude that explainability algorithms can be leveraged by adversaries and thus the advocates of training more interpretable classifiers should consider the trade-off of higher vulnerability of those classifiers to adversarial attacks.

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。

↑