Abstract
As the proliferation and adoption of mobile phones becomes more and more common, so do the targets for malicious hackers, especially on the widely used android platform. Even the most secure anti-malware systems do not stand a chance against some of the attacks that malware authors develop, despite using machine learning based methods to thwart such attempts. Therefore, to evaluate the vulnerability of machine learning models, we propose two attack scenarios, the first based on reinforcement learning techniques and the second on a deep learning technique, that will perturb malicious samples to appear benign. This will in turn yield greater rate of misclassification. To further distinguish these adversarial examples, we propose one defensive scenario that will help make these classification models more robust against such attacks.We will use three publicly available datasets to benchmark our models.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
