Florence Marcotte and Luca Calatroni, ERC Starting Grant Recipients 2023
- Institutional
- Research
Published on October 5, 2023
–
Updated on October 6, 2023
Dates
on the September 11, 2023
Inria Researcher Florence Marcotte and CNRS Researcher Luca Calatroni, both in joint research teams at Université Côte d'Azur, are recipients of 2023 ERC Starting Grants. This funding is part of the EU’s Horizon Europe programme, supporting early-career researchers to launch new projects and form their own teams.
Florence Marcotte is a researcher in fluid mechanics at the Inria research center at Université Côte d'Azur', in the Castor team at the J.A. Dieudonné Laboratory (Univ Côte d'Azur - CNRS). The ERC grant she has been awarded will fund a project that uses the mathematical theory of optimal control. In certain astrophysical objects, the formation of magnetic fields crucially depends on the initial conditions. Mathematically, this implies that the nonlinear equations driving their evolution can have multiple solutions, and depending on the initial magnetic disturbance, the system will spontaneously evolve towards one equilibrium rather than another: a dynamo solution, implying that the object can develop a persistent and self-sustained magnetic field, or conversely, a non-magnetic solution.
Identifying a dynamo equilibrium directly is very challenging due to the nature of these equations, and often requires a reasonably precise idea of the anticipated solution. Optimal control theory overcomes this hurdle, as demonstrated in a recent article in PRL (Physical Review Letters). The concept is to look for the initial disturbances that maximize the magnetic field energy at a specific time, and then, using conventional numerical simulation, to retroactively verify that these "initial seeds" can trigger a dynamo instability, tracking their evolution until the targeted equilibrium is reached.
Florence and her team aim to apply this methodology to model the radiative zones of stars, where possible dynamo mechanisms remain a mystery. For instance, an intriguing problem is the existence of the "magnetic desert" in intermediate-mass stars: the detected magnetic fields on these stars' surfaces are either extremely intense or very weak, and the reason behind this is unknown. It's a compelling challenge to try to explain how similar objects can undergo two distinct magnetic evolutions. The team will also test their approach on the question of the origin of the turbulence in protoplanetary discs (discs of gas and dust that are forming planets).
Luca Calatroni is a CNRS Researcher at I3S (Sophia Antipolis Laboratory for Computer Science, Signals and Systems), working within the Morpheme team on problems at the interface between mathematics, signal processing, and fields such as biology and archaeology/art history. He studied applied mathematics at the University of Pavia in Italy before moving to England for his Ph.D. in mathematical models for image processing at the University of Cambridge. He was a Marie Skłodowska-Curie Postdoc at the University of Genoa in Italy and a Hadamard Lecturer in the Centre des Mathématiques Appliquées at the École Polytechnique in Paris. In 2019, he joined I3S as a CNRS Researcher.
Luca’s research is focused on the mathematical modeling and numerical resolution of inverse imaging problems across various applications like biology, heritage imaging, and computational neuroscience. The ERC StG MALIN project (Model-aware learning for inverse imaging problems in fluorescence microscopy) aims at the theoretical study and application of physics-inspired learning models to several inverse problems in fluorescence microscopy. A significant challenge in this field is the super-resolution of images, essential for a deeper understanding of biological processes at the nanometer scale.
The integration of physical models into learning techniques has gained popularity in recent years, but its implementation in the field of fluorescence microscopy faces multiple challenges due to the nonlinearity of phenomena characterizing the acquisition processes or the scarcity of learning data. Crucial issues like the stability of the reconstruction call for detailed study, which will be the subject of dedicated work packages.
The methodologies developed in the project will be used to address various important questions in biology, such as the reproductive mechanisms of certain algae, in collaboration with S. Schaub (CNRS) from Villefranche-sur-Mer Developmental Biology Laboratory (LBDV).
Identifying a dynamo equilibrium directly is very challenging due to the nature of these equations, and often requires a reasonably precise idea of the anticipated solution. Optimal control theory overcomes this hurdle, as demonstrated in a recent article in PRL (Physical Review Letters). The concept is to look for the initial disturbances that maximize the magnetic field energy at a specific time, and then, using conventional numerical simulation, to retroactively verify that these "initial seeds" can trigger a dynamo instability, tracking their evolution until the targeted equilibrium is reached.
Florence and her team aim to apply this methodology to model the radiative zones of stars, where possible dynamo mechanisms remain a mystery. For instance, an intriguing problem is the existence of the "magnetic desert" in intermediate-mass stars: the detected magnetic fields on these stars' surfaces are either extremely intense or very weak, and the reason behind this is unknown. It's a compelling challenge to try to explain how similar objects can undergo two distinct magnetic evolutions. The team will also test their approach on the question of the origin of the turbulence in protoplanetary discs (discs of gas and dust that are forming planets).
Luca Calatroni is a CNRS Researcher at I3S (Sophia Antipolis Laboratory for Computer Science, Signals and Systems), working within the Morpheme team on problems at the interface between mathematics, signal processing, and fields such as biology and archaeology/art history. He studied applied mathematics at the University of Pavia in Italy before moving to England for his Ph.D. in mathematical models for image processing at the University of Cambridge. He was a Marie Skłodowska-Curie Postdoc at the University of Genoa in Italy and a Hadamard Lecturer in the Centre des Mathématiques Appliquées at the École Polytechnique in Paris. In 2019, he joined I3S as a CNRS Researcher.
Luca’s research is focused on the mathematical modeling and numerical resolution of inverse imaging problems across various applications like biology, heritage imaging, and computational neuroscience. The ERC StG MALIN project (Model-aware learning for inverse imaging problems in fluorescence microscopy) aims at the theoretical study and application of physics-inspired learning models to several inverse problems in fluorescence microscopy. A significant challenge in this field is the super-resolution of images, essential for a deeper understanding of biological processes at the nanometer scale.
The integration of physical models into learning techniques has gained popularity in recent years, but its implementation in the field of fluorescence microscopy faces multiple challenges due to the nonlinearity of phenomena characterizing the acquisition processes or the scarcity of learning data. Crucial issues like the stability of the reconstruction call for detailed study, which will be the subject of dedicated work packages.
The methodologies developed in the project will be used to address various important questions in biology, such as the reproductive mechanisms of certain algae, in collaboration with S. Schaub (CNRS) from Villefranche-sur-Mer Developmental Biology Laboratory (LBDV).
