FARADAI

Frugal and Robust AI for Defence Advanced Intelligence.

The project “Frugal and Robust AI for Defence Advanced Intelligence” (FaRADAI) focusses on frugal learning, i.e. the ability of a system to adapt and learn from its’ environment, including from user supervision, for a reasonable cost and without intervention from expert developers. An important cross- cutting need for Artificial Intelligence is to create technologies for reliable, autonomous and frugal learning, i.e. the ability of a system to adapt and learn from its environment, including from user supervision, at a reasonable cost and without ‘ intervention of expert developers nor regression. Such technologies can be highly disruptive and have a high impact on many capabilities, especially when the information to be managed is highly variable or unpredictable and high adaptability is required.

These technologies can also alleviate the current need to provide data to system developers to achieve improvements depending on that data, which can be critical when the data is sensitive, and is therefore critical for defense. They can improve technological independence more generally. Selected actions should include
the organization of technology challenges that address well- defined objectives in order to initiate and drive progress towards addressing identified defense needs, while simultaneously leveraging civilian research and generating spillover effects. As part of the FaRADAI project, current advances in artificial intelligence technologies will be thoroughly studied in parallel with a detailed study of the main challenges imposed by a defense system. Aiming for significant breakthroughs in artificial intelligence, the models will accelerate their wider application and deployment in defense systems, increasing their impact and overall performance.