Leonardo Petrini

Joined

April 2021

Names

Leonardo Petrini (Preferred)

Emails

****@epfl.ch (Confirmed)

Personal Links

Homepage

Google Scholar

DBLP

Career & Education History

PhD student

Swiss Federal Institute of Technology Lausanne (epfl.ch)

2019 – 2023

MS student

Swiss Federal Institute of Technology Lausanne (epfl.ch)

2017 – 2019

Undergrad student

Politecnico di Torino (polito.it)

2014 – 2017

Advisors, Relations & Conflicts

Coauthor

Umberto Tomasini

2022 – Present

Coauthor

Francesco Cagnetta

2021 – Present

Coauthor

Alessandro Favero

2020 – Present

PhD Advisor

Matthieu Wyart

2019 – Present

Coauthor

Mario Geiger

2019 – Present

Expertise

deep learning

2017 – Present

Publications

How Deep Convolutional Neural Networks lose Spatial Information with training
Umberto Maria Tomasini, Leonardo Petrini, Francesco Cagnetta, Matthieu Wyart
- Published: 03 Mar 2023, Last Modified: 15 Feb 2026
- Physics4ML Poster
- Readers: Everyone
How deep convolutional neural networks lose spatial information with training
Umberto Maria Tomasini, Leonardo Petrini, Francesco Cagnetta, Matthieu Wyart
- Published: 01 Feb 2023, Last Modified: 14 Jan 2026
- Submitted to ICLR 2023
- Readers: Everyone
Learning sparse features can lead to overfitting in neural networks
Leonardo Petrini, Francesco Cagnetta, Eric Vanden-Eijnden, Matthieu Wyart
- Published: 31 Oct 2022, Last Modified: 08 Feb 2026
- NeurIPS 2022 Accept
- Readers: Everyone
How deep convolutional neural networks lose spatial information with training
Umberto M. Tomasini, Leonardo Petrini, Francesco Cagnetta, Matthieu Wyart
- 2022 (modified: 16 Apr 2023)
- CoRR 2022
- Readers: Everyone
Learning sparse features can lead to overfitting in neural networks
Leonardo Petrini, Francesco Cagnetta, Eric Vanden-Eijnden, Matthieu Wyart
- 2022 (modified: 16 Apr 2023)
- CoRR 2022
- Readers: Everyone
Relative stability toward diffeomorphisms indicates performance in deep nets
Leonardo Petrini, Alessandro Favero, Mario Geiger, Matthieu Wyart
- Published: 09 Nov 2021, Last Modified: 26 May 2025
- NeurIPS 2021 Poster
- Readers: Everyone
Relative stability toward diffeomorphisms indicates performance in deep nets
Leonardo Petrini, Alessandro Favero, Mario Geiger, Matthieu Wyart
- 2021 (modified: 16 Apr 2023)
- NeurIPS 2021
- Readers: Everyone
Relative stability toward diffeomorphisms in deep nets indicates performance
Leonardo Petrini, Alessandro Favero, Mario Geiger, Matthieu Wyart
- 2021 (modified: 16 Apr 2023)
- CoRR 2021
- Readers: Everyone
Compressing invariant manifolds in neural nets
Jonas Paccolat, Leonardo Petrini, Mario Geiger, Kevin Tyloo, Matthieu Wyart
- 2020 (modified: 16 Apr 2023)
- CoRR 2020
- Readers: Everyone
Perspective: A Phase Diagram for Deep Learning unifying Jamming, Feature Learning and Lazy Training
Mario Geiger, Leonardo Petrini, Matthieu Wyart
- 2020 (modified: 16 Apr 2023)
- CoRR 2020
- Readers: Everyone

Leonardo Petrini

Names

Emails

Personal Links

Career & Education History

Advisors, Relations & Conflicts

Expertise

Publications

Co-Authors