| 基于深度强化学习的恶意软件混淆对抗样本生成 |
| |
| 引用本文： | 严莹子,王小平,庄葛巍,顾臻,贺青,史扬.基于深度强化学习的恶意软件混淆对抗样本生成[J].计算机应用与软件,2022(2):315-323+349. |
| |
| 作者姓名： | 严莹子 王小平 庄葛巍 顾臻 贺青 史扬 |
| |
| 作者单位： | 1. 同济大学电子与信息工程学院;2. 国网上海市电力公司电力科学研究院;3. 同济大学软件学院 |
| |
| 基金项目： | 国家自然科学基金项目(61772371)； |
| |
| 摘 要： | 设计一种PE格式恶意软件混淆对抗样本生成模型。利用深度强化学习算法,实现对恶意软件的自动混淆。通过加入历史帧和LSTM神经网络结构的方法使深度强化学习模型具有记忆性。对比实验表明,该恶意软件变种在基于机器学习的检测模型上的逃逸率高于现有研究,在由918个PE格式恶意软件组成的测试集上达到39.54%的逃逸率。
|
| 关 键 词： | 深度强化学习 代码混淆 对抗训练 恶意软件检测 |
OBFUSCATED CODE ADVERSARIAL SAMPLE GENERATION METHOD BASED ON DEEP REINFORCEMENT LEARNING |
| |
| Yan Yingzi,Wang Xiaoping,Zhuang Gewei,Gu Zhen,He Qing,Shi Yang.OBFUSCATED CODE ADVERSARIAL SAMPLE GENERATION METHOD BASED ON DEEP REINFORCEMENT LEARNING[J].Computer Applications and Software,2022(2):315-323+349. |
| |
| Authors: | Yan Yingzi Wang Xiaoping Zhuang Gewei Gu Zhen He Qing Shi Yang |
| |
| Affiliation: | (School of Electronic and Information Engineering,Tongji University,Shanghai 200092,China;Power Science Research Institute of Shanghai Electric Power Company,Shanghai 200436,China;School of Software Engineering,Tongji University,Shanghai 200092,China) |
| |
| Abstract: | This paper designs a PE malware obfuscation adversarial sample generation model.It used deep reinforcement learning algorithms to realize automatic obfuscation of malware,and the deep reinforcement learning model was provided with memory through historical frames and LSTM neural network structure.Comparative experiments show that the escape rate of the generated malware variants on the machine learning-based detection model is higher than that of existing research,reaching 39.54%on the test set composed of 918 PE malware. |
| |
| Keywords: | Deep reinforcement learning Code obfuscation Adversarial training Malware detection |
| 本文献已被 维普 等数据库收录！ |
|
