Pierrick Leroy

Ph.D. candidate in Intelligenza Artificiale , 39th cycle (2023-2026)
Department of Mathematical Sciences (DISMA)

Profile

PhD

Research topic

Geometric and topological methods for AI

Tutors

Research interests

Explainable ML
Generative models and Generative Adversarial Networks
Knowledge representation and reasoning

Biography

Hello! I’m Pierrick, a PhD student supervised by Prof. Francesco Vaccarino and Prof. Giovanni Petri. My current research focuses on topological data analysis tools for machine and especially deep learning. We aim to extract topological and geometric information, either from data represented internally by trained models or from the space of parameters learned during the training process. In particular, we are interested in understanding how topology and geometry intertwine with the process of learning via TDA and Morse theory. By doing so, we hope to shed some light on the inner workings of deep learning.

Research

Research groups

Publications

Latest publications View all publications in Porto@Iris

  • Vaccarino, Francesco; Nurisso, Marco; Leroy, Pierrick; Mastropietro, Antonio (2025)
    Attributes Shape the Embedding Space of Face Recognition Models. In: International Conference on Machine Learning, 13-19 July 2025, Vancouver Convention Center, Vancouver, Canada, Vancouver, Canada, 13-19 July 2025, pp. 33960-33983. ISSN 2640-3498
    Contributo in Atti di Convegno (Proceeding)
  • Nurisso, Marco; Leroy, Pierrick; Vaccarino, Francesco (2024)
    Topological obstruction to the training of shallow ReLU neural networks. In: NeurIPS 2024: Thirty-Eighth Annual Conference on Neural Information Processing Systems, Vancouver (CA), Dec 10 , 2024 - Dec 15, 2024, pp. 35358-35387. ISBN: 9798331314385
    Contributo in Atti di Convegno (Proceeding)