Causally Reliable Concept Bottleneck Models
I am a postdoctoral researcher at the Faculty of Informatics at the Università della Svizzera Italiana (USI) in Lugano, Switzerland, where I work in Prof. Silvia Santini and Prof. Marc Langheinrich’s group. I hold a Ph.D. in Computer Science from the University of Liverpool (UK) and a Master’s degree in experimental particle physics from the University of Pisa (IT), having conducted my thesis work within the Mu2e experiment at Fermi National Laboratory (US).
My research centers on interpretability-by-design: building deep learning models whose reasoning is interpretable by construction rather than explained after training, which makes them steerable and verifiable. To this end, I work on standardizing the underlying theory and code interfaces and on adding structure to the interpretable intermediate space.
Among data modalities, I focus particularly on time series, including forecasting, classification, representation learning, and virtual sensing, which I study through problems in the natural sciences and physiological signals. I am also a co-creator of the open-source machine learning library PyTorch Concepts.
