Matthieu Wyart

Publication Activity (10 Years)

Years Active: 2010-2024
Publications (10 Years): 32

Top Topics

Hierarchical Data

Learning Curves

Top Venues

Mach. Learn. Sci. Technol.

Publications

Antonio Sclocchi, Alessandro Favero, Matthieu Wyart
A Phase Transition in Diffusion Models Reveals the Hierarchical Nature of Data. CoRR (2024)
Umberto M. Tomasini, Matthieu Wyart
How Deep Networks Learn Sparse and Hierarchical Data: the Sparse Random Hierarchy Model. CoRR (2024)
Francesco Cagnetta, Matthieu Wyart
Towards a theory of how the structure of language is acquired by deep neural networks. CoRR (2024)
Antonio Sclocchi, Mario Geiger, Matthieu Wyart
Dissecting the Effects of SGD Noise in Distinct Regimes of Deep Learning. ICML (2023)
Umberto M. Tomasini, Leonardo Petrini, Francesco Cagnetta, Matthieu Wyart
How deep convolutional neural networks lose spatial information with training. Mach. Learn. Sci. Technol. 4 (4) (2023)
Francesco Cagnetta, Alessandro Favero, Matthieu Wyart
What Can Be Learnt With Wide Convolutional Neural Networks? ICML (2023)
Leonardo Petrini, Francesco Cagnetta, Umberto M. Tomasini, Alessandro Favero, Matthieu Wyart
How Deep Neural Networks Learn Compositional Data: The Random Hierarchy Model. CoRR (2023)
Antonio Sclocchi, Matthieu Wyart
On the different regimes of Stochastic Gradient Descent. CoRR (2023)
Antonio Sclocchi, Mario Geiger, Matthieu Wyart
Dissecting the Effects of SGD Noise in Distinct Regimes of Deep Learning. CoRR (2023)
Umberto M. Tomasini, Leonardo Petrini, Francesco Cagnetta, Matthieu Wyart
How deep convolutional neural networks lose spatial information with training. CoRR (2022)
Umberto M. Tomasini, Antonio Sclocchi, Matthieu Wyart
Failure and success of the spectral bias prediction for Laplace Kernel Ridge Regression: the case of low-dimensional data. ICML (2022)
Leonardo Petrini, Francesco Cagnetta, Eric Vanden-Eijnden, Matthieu Wyart
Learning sparse features can lead to overfitting in neural networks. NeurIPS (2022)
Francesco Cagnetta, Alessandro Favero, Matthieu Wyart
How Wide Convolutional Neural Networks Learn Hierarchical Tasks. CoRR (2022)
Leonardo Petrini, Francesco Cagnetta, Eric Vanden-Eijnden, Matthieu Wyart
Learning sparse features can lead to overfitting in neural networks. CoRR (2022)
Umberto M. Tomasini, Antonio Sclocchi, Matthieu Wyart
Failure and success of the spectral bias prediction for Kernel Ridge Regression: the case of low-dimensional data. CoRR (2022)
Alessandro Favero, Francesco Cagnetta, Matthieu Wyart
Locality defeats the curse of dimensionality in convolutional teacher-student scenarios. CoRR (2021)
Mario Geiger, Christophe Eloy, Matthieu Wyart
How memory architecture affects performance and learning in simple POMDPs. CoRR (2021)
Leonardo Petrini, Alessandro Favero, Mario Geiger, Matthieu Wyart
Relative stability toward diffeomorphisms in deep nets indicates performance. CoRR (2021)
Leonardo Petrini, Alessandro Favero, Mario Geiger, Matthieu Wyart
Relative stability toward diffeomorphisms indicates performance in deep nets. NeurIPS (2021)
Alessandro Favero, Francesco Cagnetta, Matthieu Wyart
Locality defeats the curse of dimensionality in convolutional teacher-student scenarios. NeurIPS (2021)
Jonas Paccolat, Stefano Spigler, Matthieu Wyart
How isotropic kernels perform on simple invariants. Mach. Learn. Sci. Technol. 2 (2) (2021)
Jonas Paccolat, Stefano Spigler, Matthieu Wyart
How isotropic kernels learn simple invariants. CoRR (2020)
Barbara Bravi, Riccardo Ravasio, Carolina Brito, Matthieu Wyart
Direct coupling analysis of epistasis in allosteric materials. PLoS Comput. Biol. 16 (3) (2020)
Jonas Paccolat, Leonardo Petrini, Mario Geiger, Kevin Tyloo, Matthieu Wyart
Compressing invariant manifolds in neural nets. CoRR (2020)
Mario Geiger, Leonardo Petrini, Matthieu Wyart
Perspective: A Phase Diagram for Deep Learning unifying Jamming, Feature Learning and Lazy Training. CoRR (2020)
Mario Geiger, Arthur Jacot, Stefano Spigler, Franck Gabriel, Levent Sagun, Stéphane d'Ascoli, Giulio Biroli, Clément Hongler, Matthieu Wyart
Scaling description of generalization with number of parameters in deep learning. CoRR (2019)
Stefano Spigler, Mario Geiger, Matthieu Wyart
Asymptotic learning curves of kernel methods: empirical data v.s. Teacher-Student paradigm. CoRR (2019)
Mario Geiger, Stefano Spigler, Arthur Jacot, Matthieu Wyart
Disentangling feature and lazy learning in deep neural networks: an empirical study. CoRR (2019)
Stefano Spigler, Mario Geiger, Stéphane d'Ascoli, Levent Sagun, Giulio Biroli, Matthieu Wyart
A jamming transition from under- to over-parametrization affects loss landscape and generalization. CoRR (2018)
Marco Baity-Jesi, Levent Sagun, Mario Geiger, Stefano Spigler, Gérard Ben Arous, Chiara Cammarota, Yann LeCun, Matthieu Wyart, Giulio Biroli
Comparing Dynamics: Deep Neural Networks versus Glassy Systems. CoRR (2018)
Mario Geiger, Stefano Spigler, Stéphane d'Ascoli, Levent Sagun, Marco Baity-Jesi, Giulio Biroli, Matthieu Wyart
The jamming transition as a paradigm to understand the loss landscape of deep neural networks. CoRR (2018)
Marco Baity-Jesi, Levent Sagun, Mario Geiger, Stefano Spigler, Gérard Ben Arous, Chiara Cammarota, Yann LeCun, Matthieu Wyart, Giulio Biroli
Comparing Dynamics: Deep Neural Networks versus Glassy Systems. ICML (2018)
Edan Lerner, Gustavo Düring, Matthieu Wyart
Simulations of driven overdamped frictionless hard spheres. Comput. Phys. Commun. 184 (3) (2013)
Matthieu Wyart, David Botstein, Ned S. Wingreen
Evaluating Gene Expression Dynamics Using Pairwise RNA FISH Data. PLoS Comput. Biol. 6 (11) (2010)