Alessandro Favero

I’m an inaugural Physics-AI Fellow at the University of Cambridge.

My research aims to advance the fundamental understanding of artificial intelligence, often drawing on methods from statistical physics and complex systems theory. A primary focus of my work is to characterize the latent structure of data, such as language and images, and to understand how generative models build internal representations of this structure. I’m also interested in leveraging these foundational insights to accelerate scientific discovery.

I obtained my PhD from EPFL (Switzerland), where I worked on the physics of learning and AI under the supervision of Matthieu Wyart and Pascal Frossard. My academic background is in the physics of complex systems, with studies at Sorbonne, Politecnico di Torino, SISSA, and ICTP. I’ve also spent time as an Applied Scientist Intern with the Fundamental Research team at Amazon’s AWS AI Labs in the Bay Area.