Matching AI Adversaries to Pilot Trainees

AI pilots offer the potential of being tireless trainers that are always available to rigorously train to the curriculum. AI agents are expected to augment expert human trainers who are in limited supply and unavailable on-demand.
 
AFRL and Aptima are working on a system to smartly match AI adversaries to pilot trainees to improve access to and quality of simulated training. A unique feature of this new AI system will be to leverage training data in real-time. As performance measures roll in, an adaptive learning model will select the next scenario and agent that will advance the trainee to expert-level most quickly and surely. Rather than apply a traditional, one-size-fits-all approach, these models personalize learning at scale.