MALICE

The goal of the MALICE project-team is to combine interdisciplinary skills in statistical learning and laser-matter interaction to foster the development of new joint methodological contributions at the interface between Machine Learning and Surface Engineering. The members of MALICE have complementary backgrounds in computer science, applied mathematics, statistics and optimization. They also benefit from the expertise of physicists of the Hubert Curien lab in modeling ultrashort laser-matter interaction which makes possible scientific breakthroughs in both domains. On the one hand, surface engineering raises numerous machine learning challenges, including (i) a limited access to training data and the availability of only partial and incomplete background knowledge (typically in the form of non linear PDEs), (ii) the need of deriving theoretical guarantees for Physics-informed learning models trained from both data and physical knowledge and (iii) a strong necessity to transfer knowledge from one learned dynamical system to another. On the other hand, the advances carried out in machine learning in MALICE allow to better understand the physics underlying the mechanisms of laser/radiation-matter interaction, enabling to address numerous societal challenges in the fields of space, nuclear, defense, energy or health.