Main Challenges for Machine Learning in Critical Systems

Grégory FLANDIN
PhD, Program Director at IRT Saint Exupéry
November 9th, 2023

Abstract
Recent progress in Artificial Intelligence, especially in Machine Learning, has aroused unprecedented interest in these technologies. Many industrial sectors are now considering using them. However, this has led to strong scientific obstacles. Machine learning, especially deep neural networks, can perform well enough to consider critical applications such as autonomous vehicles, predictive maintenance and medical diagnosis, but their theoretical properties are not well-known yet. These scientific challenges make it difficult to meet the industrial constraints required for a general application such as certification, qualification and explainability of algorithms. In the frame of the DEEL project, research teams in Quebec and France are working since 2018 on the development of dependable, robust and explainable artificial intelligence. The talk will review some of the major challenges and provide parts of solutions that can be envisaged from the most recent scientific and technological developments.