The research focuses on Reinforcement Learning, a field of Artificial Intelligence that enables systems to make sequential decisions, much like in a chess game. While recent advances driven by Deep Neural Networks have achieved remarkable results, these solutions often lack interpretability, making it difficult to understand and explain the reasoning behind their strategic decisions.

To address this limitation, the study by Selmonaj and Antonucci aims to enhance the interpretability of aircraft control systems in military settings, where understanding the rationale behind critical decisions is crucial. The research specifically explores multi-agent systems, in which an intelligent system manages an entire fleet of aircraft, adding to the interpretability challenges.

This project is the outcome of a long-standing collaboration between SUPSI’s Dalle Molle Institute for Artificial Intelligence and the DDPS, dedicated to developing trustworthy AI-based defense technologies—ethical, transparent, and reliable solutions.

The Best Paper Award was presented during the Annual Symposium Modeling and Simulation as an Enabler for Digital Transformation in NATO and Nations, organized by the Science and Technology Organization (STO), NATO's science and technology body. This recognition underscores the importance of “explainability” in artificial intelligence as a key element for the future of defense and security technologies.

The award highlights the growing importance of explainability in artificial intelligence as a fundamental element for the future of defense and security technologies.