Many natural and engineered systems, from turbulent fluids to traffic flows, are described by complex equations that are often too difficult to derive or compute in practice.

During his PhD studies, Simone Manti developed new machine learning methods that automatically discover governing equations from data. Unlike traditional approaches, these methods produce interpretable mathematical formulas, offering new insights into the underlying physics.

At the core of the work is symbolic regression, a technique that searches for equations starting from a pool of simple functions rather than fitting data to pre-defined models. By combining symbolic regression with discrete mathematics, Simone Manti created a framework capable of discovering new models for systems governed by partial differential equations.

These approaches were applied to fluid transport problems, from reduced-order to traffic flow modeling, demonstrating how data-driven discovery can improve predictions while keeping the models transparent and physically meaningful. The research also produced two open-source software libraries from the ground up, potentially helping other scientists explore and extend these ideas.

The PhD study was completed at the Department of Mechanical and Production Engineering, Faculty of Technical Sciences, Aarhus University.

This summary was prepared by the PhD student.

Time: Thursday, 20 November 2025 at 09:00
Place: Building 5510, room 104, small auditorium, INCUBA, Åbogade 15, 8200 Aarhus N.
Title of PhD thesis: Model Discovery via Symbolic Regression with Applications to Fluid Transport Problems
Contact information: Simone Manti, e-mail: smanti@mpe.au.dk, manti998@gmail.com, tel.: +45 60 20 26 57
Members of the assessment committee:
Associate Professor Giovanni Stabile, The BioRobotics Institute, Sant’Anna School of Advanced Studies, Italy
Associate Professor Francesco Regazzoni, Department of Mathematics, Politecnico di Milano, Italy
Associate Professor Michal K. Budzik (chair), Department of Mechanical and Production Engineering, Aarhus University, Denmark
Main supervisor: Associate Professor Alessandro Lucantonio, Department of Mechanical and Production Engineering, Aarhus University, Denmark
Language: The PhD dissertation will be defended in English

The defence is public.
The PhD thesis is available for reading at the Graduate School of Technical Sciences/GSTS, Ny Munkegade 120, building 1521, 8000 Aarhus C.