EASA and Daedalean created Concepts of Design Assurance for Neural Networks

Together with European Union Aviation Safety Agency (EASA), we have concluded the first project to address the challenges of using Neural Networks and Machine Learning in safety-critical aviation. The public extract of the project report, entitled “Concepts of Design Assurance for Neural Networks” is published on the EASA website. This project is a first major step in the definition of the “Learning Assurance” process and an enabler towards full certification and approval of machine learning applications on-board aircraft – a requirement for Urban Air Mobility and beyond. Thanks to Guillaume Soudain, Xavier Henriquel, Georges Rebender and Jean Marc Cluzeau for the collaboration and expertise, and to David Haber and other members of our team for the hard work! More work needs to be done and we are looking forward to continuing working with EASA to lead the way in thinking about trustworthy AI. Read our full press release to learn the details