The U.S. Air Force is arguably the best-trained air force in the world. But this comes at an annual price of ~$750M. The Air Force has continuing interest in reducing training costs while maintaining or improving the training product. Thus, there is tremendous need for educational approaches that would customize training to the individual, minimizing the trainee's time in training by focusing on the trainee's needs, and getting the trainee productive sooner, leading to reduced costs. The Air Education and Training Command serves about 800,000 users annually who complete up to 1M courses monthly, providing a centralized data collection and distribution point for the core learning services. Yet, there is no integrated intelligent agent that continuously learns from expert behavior. nFlux leverages Sigma (ã) cognitive architecture to offer a set of control services (ã-CS) that imitate expert behavior such that Airmen can learn complex behavior sets to perform at optimum levels. ã-CS adapts to the trainee's needs to personalize the training experience, minimizing the need for professional educators. ã-CS takes an Imitation Learning approach to accelerate the Air Force's capability to enable, e.g., Pilots, to imitate expert behavior and continuously improve based on cognitive learning through an intelligent
