Laura’s PhD research applies machine learning to large-scale sensor data from the ATLAS experiment at CERN. She developed a neural network method to calibrate particle energy measurements, improving accuracy and reducing the need for simulation.
She also implemented and evaluated transformer-based models for anomaly detection in multidimensional time series, using both real and simulated detector data in collaboration with IBM Research. The methods she developed for data calibration and fault detection are broadly applicable to industrial monitoring, IoT systems, and other data-intensive environments.
