Adversarial genetic programming for cyber security: a rising application domain where GP matters

Abstract

Cyber security adversaries and engagements are ubiquitous and ceaseless. We delineate Adversarial Genetic Programming for Cyber Security, a research topic that, by means of genetic programming (GP), replicates and studies the behavior of cyber adversaries and the dynamics of their engagements. Adversarial Genetic Programming for Cyber Security encompasses extant and immediate research efforts in a vital problem domain, arguably occupying a position at the frontier where GP matters. Additionally, it prompts research questions around evolving complex behavior by expressing different abstractions with GP and opportunities to reconnect to the Machine Learning, Artificial Life, Agent-Based Modeling and Cyber Security communities. We present a framework called RIVALS which supports the study of network security arms races. Its goal is to elucidate the dynamics of cyber networks under attack by computationally modeling and simulating them.

Coevolutionary modeling of cyber attack patterns and mitigations using public datasets

“…Thus we can model two populations where the members of each compete, with our formulated competition, pairwise against each other, and the fitness of each member is the sum of the harm that is either inflicted (a maximizing objective for attacks) or incurred (a minimizing objective of the same quantity: harm, for defenses) over all a member's competitions. We note that this basic formulation of interacting circumstances and many competitions also underpin prior work such as [20,22].…”

“…One basis of realism is whether a system runs actual attack code (aka malware) like a threat-actor does. On this basis, RIVALS-Deception [20] runs attack code (when it runs nmap to reconnaissance scan). In the middle of the spectrum of realism, RIVALS-DDOS [20] runs code that models a DOS attack, rather than execute DOS malware.…”

Ehrlich and Raven escape and radiate coevolution hypothesis at different levels of organization: Past and future perspectives

“…Pimentel (1961) developed the idea that reciprocal genetic changes can regulate population size in resource-consumer interactions. Even though the idea of coevolution was already there, it was Ehrlich and Raven's thought-provoking publication in 1964 that popularized the term coevolution, impacting ideas even outside biology (O'Reilly et al 2020). During those same years, Janzen's (1966) landmark studies on coevolution of plants and ants, and Smith's (1970) studies of coevolution of pines and squirrels, were published and influenced subsequent studies, especially in evolutionary ecology.…”

Software Testing or The Bugs’ Nightmare

“…This process has successfully optimised software in different ways, some examples are energy [17] and time consumption [49]. It also has already opened new areas of research such as adversarial genetic programming for cybersecurity [66]. However, the main limitation of genetic improvement is code readability, which is one of the open problems of the area [1].…”