Hacking Neural Networks: A Short Introduction
It addresses security vulnerabilities in neural networks for practitioners and researchers, though it is incremental by compiling known attacks.
The paper introduces a range of simpler security attacks on neural networks, including backdooring and GPU-based buffer overflows, and provides open-source exercises for practical exploration.
A large chunk of research on the security issues of neural networks is focused on adversarial attacks. However, there exists a vast sea of simpler attacks one can perform both against and with neural networks. In this article, we give a quick introduction on how deep learning in security works and explore the basic methods of exploitation, but also look at the offensive capabilities deep learning enabled tools provide. All presented attacks, such as backdooring, GPU-based buffer overflows or automated bug hunting, are accompanied by short open-source exercises for anyone to try out.