AUTOMATED RED TEAMING: AN OBJECTIVE-BASED DATA FARMING APPROACH FOR RED TEAMING

摘要

In this paper, we describe an objective-based Data Farming approach for red teaming called Automated Red Teaming (ART). The main idea is to develop an ART framework using Evohitionary Algorithms (EAs), Parallel Computing and Simulation, and apply it to uncover exploitable gaps in military operational concepts, complementing the Manual Red Teaming (MRT) effort. The capability of the ART framework was evaluated vis-a-vis MRT using two maritime security scenarios addressed at the International Data Farming Workshops (IDFWs) 14 and 15. The evaluation showed that, in general, results from ART were better than those obtained from MRT, some of which were non-intuitive and surprising Solutions.