Stochastic modeling of self-evolving botnets with vulnerability discovery

Machine learning techniques have been actively studied and achieved significant performance improvements in various kinds of tasks. While we benefit from such techniques in many ways, they can be a serious security threat to the Internet if malicious attackers become able to utilize them to discover unknown software vulnerabilities. This paper introduces a new concept of self-evolving botnets, where computing resources of infected hosts are exploited to discover unknown vulnerabilities in non-infected hosts and the botnets evolve autonomously. We provide a stochastic epidemic model for the self-evolving botnets, and show its behaviors through numerical and simulation experiments.