arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1912.05945cs.LGcs.ITmath.ITstat.ML
收藏

一种鲁棒分类器:基于MDL的对抗样本生成方法

Towards a Robust Classifier: An MDL-Based Method for Generating Adversarial Examples

The University of Newcastle(纽卡斯尔大学)

Behzad Asadi, Vijay Varadharajan

摘要:

我们解决了机器学习中的对抗性示例问题,其中对手试图通过对原始样本进行功能保留修改来误导分类器。我们假设一个黑盒的情况下,对手只能访问的功能集,和分类器的最终硬决策输出。我们提出了一种使用最小描述长度(MDL)原则生成对抗性示例的方法。我们的最终目标是提高分类器的鲁棒性,在重建分类器时考虑生成的例子。我们评估我们的方法在便携式可执行(PE)文件中的静态恶意软件检测的应用。我们认为PE文件的API调用作为其区别特征,其中特征向量是表示API调用的存在-不存在的二进制向量。在我们的方法中,我们首先通过查询目标分类器来创建良性样本的数据集。接下来,我们使用MDL原理构造一个频繁模式的码表,用于压缩该数据集。最后,我们通过从良性代码表中选择一个模式并将其添加到恶意软件样本中,来生成与恶意软件样本相对应的对抗性示例。所选模式是在给定代码表的情况下使压缩对抗示例的长度最小化的模式。此修改保留了原始恶意软件样本的功能,因为保留了所有原始API调用,并且仅添加了一些新的API调用。考虑到一个神经网络,我们发现对抗性示例的规避率为78.24%,而原始恶意软件样本的规避率为8.16%。这表明了我们的方法在生成重建分类器时需要考虑的示例方面的有效性。

英文摘要:

We address the problem of adversarial examples in machine learning where an adversary tries to misguide a classifier by making functionality-preserving modifications to original samples. We assume a black-box scenario where the adversary has access to only the feature set, and the final hard-decision output of the classifier. We propose a method to generate adversarial examples using the minimum description length (MDL) principle. Our final aim is to improve the robustness of the classifier by considering generated examples in rebuilding the classifier. We evaluate our method for the application of static malware detection in portable executable (PE) files. We consider API calls of PE files as their distinguishing features where the feature vector is a binary vector representing the presence-absence of API calls. In our method, we first create a dataset of benign samples by querying the target classifier. We next construct a code table of frequent patterns for the compression of this dataset using the MDL principle. We finally generate an adversarial example corresponding to a malware sample by selecting and adding a pattern from the benign code table to the malware sample. The selected pattern is the one that minimizes the length of the compressed adversarial example given the code table. This modification preserves the functionalities of the original malware sample as all original API calls are kept, and only some new API calls are added. Considering a neural network, we show that the evasion rate is 78.24 percent for adversarial examples compared to 8.16 percent for original malware samples. This shows the effectiveness of our method in generating examples that need to be considered in rebuilding the classifier.

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。

↑