Machine Learning Development for Quantum Computing and Neutrino Physics

by Annalisa De Lorenzis

Friday Jul 10, 2026, 11:00 AM → 2:00 PM Europe/Madrid

IFAE Seminar Room

Abstract: This thesis explores how machine learning can build effective representations of complex physics data. It studies Quantum Extreme Learning Machines as hybrid classical-quantum machine learning frameworks using features for image classification, investigating the role of encoding, dynamics, entanglement and classical simulability. The thesis also applies deep learning to simulated Water Cherenkov detector images, using charge and timing information to distinguish single-vertex neutrino events from pile-up events. Finally, it presents a first proof-of-concept application of QELMs to realistic neutrino detector images.

 

Supervisors: M. Pilar Casado, Arnau Riera i Thorsten Lux.