The following research projects are available within the Ph.D. Programme in High Performance Scientific Computing (a.y. 2026/2027), each associated with a funded scholarship.
HPSS – Applications to FLASH Radiotherapy
Number of positions: 2
Supervisor: Giuseppe Felici
Research topics include computational methods and models for FLASH radiotherapy, as well as radiobiological investigations of the FLASH effect.
High-Performance Scientific Computing for Quantitative Large Deviations in Deep Learning
Number of positions: 1
Supervisor: Dario Trevisan
PhD project within the OTODDLE applied research line, focusing on initialization and training of deep neural networks, quantitative estimates, HPC-oriented algorithms, and large-scale simulations, in close interaction with the research group.
Open Frameworks for Advanced Finite Element Analysis in Multiphysics and Multiscale Systems
Number of positions: 1
Supervisor: Luca Heltai
The project focuses on the development of open-source frameworks for advanced finite element analysis in multiphysics and multiscale problems, integrating mathematical models, efficient algorithms, and scalable implementations, with validation on realistic applications.
HPC Codes for Advanced Electronic Devices and AI/ML Integration
Number of positions: 1
The research activity focuses on the development, implementation, and use of high-performance computing (HPC) codes for the design, simulation, and analysis of electronic devices based on advanced and innovative materials. In particular, complex physical and mathematical models are addressed to describe charge transport phenomena, multiphysics couplings, and nonlinear behaviors typical of next-generation devices. The use of parallel computing architectures enables large-scale simulations with high accuracy and reduced computational time. To support these activities, artificial intelligence and machine learning techniques are integrated for parameter identification, model order reduction, design optimization, and analysis of simulation data, thereby improving both the efficiency and predictive capability of computational tools.