Bridging the Gap from Research to Reality: Methods for Fortifying Mitigation Measures Against Adversarial AI

While research in the area of Adversarial AI and Mitigation (AAI&M) has quickly grown, research toward investigating the physical realizability of these methods has lagged behind. AAI&M Research often focuses on constrained or 2-Dimensional scenarios that don’t have direct translation to “in-the-wild” real-world problems. As example, an adversarial patch is often only generated and subsequently re-trained against using a single static image, which does not translate to our 3D constantly-in-motion world. To bridge the significant gap between laboratory experiment and real-world scenario, AAI&M research must dive further into unconstrained, noise-and-pose-invariant approaches to AAI how to secure against it. This chapter will focus on multiple distinct AAI&M scenarios and modeling approaches that mimic the unconstrained environments of the physical world.