Measuring Vulnerabilities of Malware Detectors with Explainability-Guided Evasion Attacks
Authors:
Preprint
Abstract
Numerous open-source and commercial malware detectors are available. However, their efficacy is threatened by new adversarial attacks, whereby malware attempts to evade detection, e.g., by performing feature-space manipulation. In this work, we propose an explainability-guided and model-agnostic framework for measuring the efficacy of malware detectors when confronted with adversarial attacks. The framework introduces the concept of Accrued Malicious Magnitude (AMM) to identify which malware features should be manipulated to maximize the likelihood of evading detection. We then use this framework to test several state-of-the-art malware detectors' ability to detect manipulated malware. We find that (i) commercial antivirus engines are vulnerable to AMM-guided manipulated samples; (ii) the ability of a manipulated malware generated using one detector to evade detection by another detector (i.e., transferability) depends on the overlap of features with large AMM values between the different detectors; and (iii) AMM values effectively measure the importance of features and explain the ability to evade detection. Our findings shed light on the weaknesses of current malware detectors, as well as how they can be improved.Keywords
All Related Versions
This publication has 7 references indexed in Scilit:
- Obfuscapk: An open-source black-box obfuscation tool for Android appsSoftwareX, 2020
- A Discussion of 'Adversarial Examples Are Not Bugs, They Are Features'Distill, 2019
- How Can We Craft Large-Scale Android Malware? An Automated Poisoning Attack2019
- A Multimodal Deep Learning Method for Android Malware Detection Using Various FeaturesIEEE Transactions on Information Forensics and Security, 2018
- Yes, Machine Learning Can Be More Secure! A Case Study on Android Malware DetectionIEEE Transactions on Dependable and Secure Computing, 2017
- ICCDetector: ICC-Based Malware Detection on AndroidIEEE Transactions on Information Forensics and Security, 2016
- Stealth attacks: An extended insight into the obfuscation effects on Android malwareComputers & Security, 2015