The CEMRACS (Centre d’Eté Mathématique de Recherche Avancée en Calcul Scientifique) is a scientific event of the SMAI (the French Society of Applied and Industrial Mathematics). It was initiated in 1996 by Yvon Maday and Frédéric Coquel and takes place every year at CIRM in Luminy (Marseille, France) for six weeks.  The aim is to bring together scientists from both the academic and industrial communities to exchange ideas about a modern and relevant scientific topic.

More information on the event can be found on the webpage: http://smai.emath.fr/cemracs/cemracs21/

CEMRACS is composed of two parts. The first is a one-week summer school with lectures of leading scientists where the goal is to give an overview of the topic to non-expert participants. The remaining five weeks are dedicated to research projects proposed by academic scientists or industrial partners. Each project is carried out by a small group of students under the supervision of experienced researchers. The outcome of the research work done by the students is usually enhanced through publications in the ESAIM:Proceedings and Reviews journal. Experience from the previous years has shown the significant positive impact of the CEMRACS, not only on the development of these short-term projects, but also on the development of more long-term interactions between mathematics, physical sciences and industry. 

A significant challenge arising in increasingly many modern applications is how to blend complex mathematical models, often based on differential or integral equations, with the large and, possibly, noisy data sets which are now routinely available in many fields of engineering, science and technology. Growing efforts are made to develop a coherent mathematical framework where one of the main obstructions is the high dimensionality of the involved mathematical objects, which requires efficient sampling or optimization techniques for large-scale problems and, possibly, model order reduction strategies. The CEMRACS 2021 will be devoted to this topic of data assimilation and model reduction in high-dimensional problems. It is by nature interdisciplinary, not only by the amount of the different mathematical skills that need to be invoked, but also by the detailed knowledge that each specific application requires.

These issues are particularly relevant in molecular simulations and several research projects in the second part of the school will be dedicated to data assimilation and model-order reduction in molecular simulations applications.