Can machine learning model with static features be fooled: an adversarial machine learning approach
- PDF / 1,450,552 Bytes
- 21 Pages / 595.276 x 790.866 pts Page_size
- 31 Downloads / 288 Views
(0123456789().,-volV)(0123456789(). ,- volV)
Can machine learning model with static features be fooled: an adversarial machine learning approach Rahim Taheri1 • Reza Javidan1 • Mohammad Shojafar2 • P. Vinod2 • Mauro Conti2 Received: 8 October 2019 / Revised: 22 February 2020 / Accepted: 1 March 2020 Ó Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract The widespread adoption of smartphones dramatically increases the risk of attacks and the spread of mobile malware, especially on the Android platform. Machine learning-based solutions have been already used as a tool to supersede signature-based anti-malware systems. However, malware authors leverage features from malicious and legitimate samples to estimate statistical difference in-order to create adversarial examples. Hence, to evaluate the vulnerability of machine learning algorithms in malware detection, we propose five different attack scenarios to perturb malicious applications (apps). By doing this, the classification algorithm inappropriately fits the discriminant function on the set of data points, eventually yielding a higher misclassification rate. Further, to distinguish the adversarial examples from benign samples, we propose two defense mechanisms to counter attacks. To validate our attacks and solutions, we test our model on three different benchmark datasets. We also test our methods using various classifier algorithms and compare them with the state-of-the-art data poisoning method using the Jacobian matrix. Promising results show that generated adversarial samples can evade detection with a very high probability. Additionally, evasive variants generated by our attack models when used to harden the developed anti-malware system improves the detection rate up to 50% when using the generative adversarial network (GAN) method. Keywords Adversarial machine learning Android malware detection Poison attacks Generative adversarial network Jacobian algorithm
1 Introduction
& Mohammad Shojafar [email protected]; [email protected] Rahim Taheri [email protected] Reza Javidan [email protected] P. Vinod [email protected] Mauro Conti [email protected] 1
Computer Engineering and IT Department, Shiraz University of Technology, Shiraz, Iran
2
Department of Mathematics, University of Padua, Padua, Italy
Nowadays using the Android application is very popular on mobile platforms. Every Android application has a Jar-like APK format and is an archive file which contains Android manifest and Classes.dex files. Information about the structure of the Apps holds in the manifest file and each part is responsible for certain actions. For instance, the requested permissions must be accepted by the users for successful installation of applications. The manifest file contains a list of hardware components and permissions required by each application. Furthermore, there are environment settings in the manifest file that are useful for running applications. The compiled source code from each application is saved as the
Data Loading...
