Advanced methods for inverse problems based on differential models and data-driven approaches - 2026_IDR_DMAT_3
Scientific-Disciplinary Group
01/MATH-05 - Numerical Analysis
Description
The research program aims to develop methods for the solution of inverse problems, with particular emphasis on imaging applications, by integrating approaches at the interface of Numerical Analysis, Optimal Control, and Machine Learning. The objective is to estimate unobservable quantities from indirect measurements, which are related to them through differential models describing the underlying physics of the phenomenon. These problems will be formulated as PDE-constrained optimal control problems. Their solution will be supported by data-driven methodologies at multiple levels: for example, by restricting the solution space through the introduction of priors learned from precomputed datasets, or by accelerating the solution of the forward problem. Project funded by Ministero dell’Università e della Ricerca in the context of FIS 2, SYNERGIZE, code FIS-2023-02228, CUP D53C24005440001.
Job posting website
Number of positions
1
Funding body
Politecnico di Milano
Selection process
Click to expand
View the original posting on the MUR website: Go to MUR website