Project Description

DESIRE (Designs pour des champs spatiaux aléatoires) is a basic research project funded by  :

• the French ANR (Agence Nationale de Recherche) and
• the Austrian FWF (Fonds zur Forderung der wissenschaftlichen Forschung).

DESIRE addresses the problem of defining efficient designs for observation of spatial fields, i.e. scalar functions of several variables.

The project adopts a statistical framework, in order to capture the two-fold use of gathered information: to infer the detailed characteristics of each sampled neighborhood of the field on the one hand, and, on the other hand, to learn its global internal structure. While the problem that we assess has numerous industrial and scientific applications, the work of the project stays at a fundamental level, with the aim of identifying appropriate modeling frameworks for distinct specific instances of the spatial observation problem, as well as expedite algorithms to compute associated efficient designs.

To demonstrate the relevance of the results obtained, and in cooperation with external collaborators of the institutions involved, the project also compares the techniques developed to existing state-of-the-art approaches in a small number of real problems.

The project work is organized in 4 core technical tasks, together with the application Task 5, where problems identified with our external collaborator MUMM are studied.