Project Lead: Dr. Matt Turek
Sponsoring Organization: DARPA
Project Synopsis: Autonomous systems that perceive, learn, decide, and act on their own are on the horizon; however, their current inability to explain their decisions and actions to human users will limit their effectiveness. If future warfighters are to understand, appropriately trust, and effectively manage an emerging generation of artificially intelligent machine partners, explainable AI is essential. XAI aims to create a suite of machine learning techniques that produce more explainable models, while maintaining a high level of learning performance (prediction accuracy). The goal is to generate a portfolio of methods that will give future developers a range of design options that cover the performance-versus-explainability trade space.