Francesco Cagnetta
École Polytechnique Fédérale de Lausanne
H-index: 8
Europe-Switzerland
Top articles of Francesco Cagnetta
How deep convolutional neural networks lose spatial information with training
Machine Learning: Science and Technology
2023/11/9
Kernels, Data & Physics
arXiv preprint arXiv:2307.02693
2023/7/5
Francesco Cagnetta
H-Index: 6
Julia Kempe
H-Index: 24
How Deep Neural Networks Learn Compositional Data: The Random Hierarchy Model
arXiv e-prints
2023/7
Francesco Cagnetta
H-Index: 6
Leonardo Petrini
H-Index: 1
Alessandro Favero
H-Index: 0
Matthieu Wyart
H-Index: 41
Feature learning and overfitting in neural networks
Bulletin of the American Physical Society
2023/3/6
Francesco Cagnetta
H-Index: 6
Statistical Mechanics of Infinitely-Wide Convolutional Networks
Bulletin of the American Physical Society
2023/3/6
Learning sparse features can lead to overfitting in neural networks
Journal of Statistical Mechanics: Theory and Experiment
2023/11/15
What can be learnt with wide convolutional neural networks?
International Conference on Machine Learning, 2023
2022/8/1
Renormalization group study of the dynamics of active membranes: Universality classes and scaling laws
Physical Review E
2022/1/27
Francesco Cagnetta
H-Index: 6
Universal properties of active membranes
Physical Review E
2022/1/27
Francesco Cagnetta
H-Index: 6
Work fluctuations in the active Ornstein–Uhlenbeck particle model
Journal of Statistical Mechanics: Theory and Experiment
2021/12/23
Antonio Suma
H-Index: 3
Francesco Cagnetta
H-Index: 6
Locality defeats the curse of dimensionality in convolutional teacher-student scenarios
Advances in Neural Information Processing Systems
2021/12/6
Work fluctuations of self-propelled particles in the phase separated state
Journal of Physics A: Mathematical and Theoretical
2020/8/14
Active interfaces, a universal approach
2020/6/25
Francesco Cagnetta
H-Index: 6
Nonequilibrium Strategy for Fast Target Search on the Genome
Physical Review Letters
2020/5/12
Kinetic roughening in active interfaces
EPJ Web of Conferences
2020/3/11
Francesco Cagnetta
H-Index: 6
Efficiency of one-dimensional active transport conditioned on motility
Physical Review E
2020/2/26
Co-Authors
