Luca Pesce

Luca is a postdoctoral research fellow at Harvard Center for Mathematical Sciences and Applications (CMSA).

His research investigates the theoretical foundations of deep learning systems, with the goal of understanding why they are so effective in practice. He studies how general-purpose training methods exploit the low-dimensional latent structure hidden in high-dimensional data, drawing on tools from statistical physics and high-dimensional statistics.

He received his PhD from the École polytechnique fédérale de Lausanne (EPFL), where he was advised by Florent Krzakala. During his PhD spent a few months as a research engineer intern working on generative algorithms for structure-based drug discovery. Earlier, he took part in an International MSc program in Theoretical Physics delocalized in Paris, Trieste, and Turin.