My research
I study fundamental mathematical and algorithmic aspects of intelligent systems, with a focus on efficient neural networks, graph neural networks, transformer-based neural solvers, multi-agent computation, and brain-inspired models of intelligence. My work spans machine learning, computational neuroscience, and distributed systems.
One line of my work has studied when large neural networks contain sparse subnetworks that preserve target behavior. This connects the expressive limits of pruning, approximation guarantees, structured architectures, and quantization. The goal is to understand which compression methods preserve capability and which impose fundamental costs.
I give a quick overview of my contributions here.
About me
I am a CNRS researcher (HDR) at Université Côte d'Azur, a 3IA Côte d'Azur Chair, a member of the COATI Project-Team between INRIA and I3S, and the coordinator of the ELECTRON Associated INRIA Team with King's College London. Currently, I am a long-term visitor in Prof. Petra Mutzel's group at University of Bonn and a member of the Lamarr Institute.
https://orcid.org/0000-0002-8755-3892Short Bio
Before accepting a CNRS position in 2018, I've been a fellow of the Simons Institute for the Theory of Computing in the Brain and Computation Program and a postdoctoral fellow at the Max Planck Institute for Informatics. In 2019, I've received the Best Italian Young Researcher in Theoretical Computer Science award by the Italian Chapter of the European Association of Theoretical Computer Science, from which I also received the Best PhD Thesis in Theoretical Computer Science in 2017. In 2016, I've been a recipient of the Best Student Paper Award at the European Symposium on Algorithms. I have led two national research projects funded by the French Directorate General of Armament: BraInside (2019-23), which focused on neural network pruning, and BioSwarm (2024-27), which focused on bio-inspired strategies for drone swarms. For more information (projects, academic service etc.), see my CV.