Adversarial Malware Generation Using Binary Malware Samples
A Comparison of State-of-the-Art Techniques for Generating Adversarial Malware Binaries
We consider the problem of generating adversarial malware by a cyber-attacker where the attacker s task is to strategically modify certain bytes within existing binary malware files, so that the modified files are able to evade a machine learning-based malware classifier such as machine learning-based malware classifier.
We evaluate three recent adversarial malware generation techniques using binary malware samples drawn from a single, publicly available malware data set and compare their performances for evading a machine-learning based malware classifier called malconv.
Our results show that among the compared techniques, the most effective technique is the one that strategically modifies bytes in abinary s header.