Research

Theory of deep learning

A Phase Transition in Diffusion Models Reveals the Hierarchical Nature of Data, Antonio Sclocchi, Alessandro Favero, Matthieu Wyart, arxiv:2402.16991 (2024).

On the different regimes of stochastic gradient descent, Antonio Sclocchi, Matthieu Wyart, Proceedings of the National Academy of Sciences, 121 (9), e2316301121 (2024).

How Deep Neural Networks Learn Compositional Data: The Random Hierarchy Model, Leonardo Petrini, Francesco Cagnetta, Umberto Tomasini, Alessandro Favero, Matthieu Wyart, arXiv:2307.02129 (2023).

Dissecting the Effects of SGD Noise in Distinct Regimes of Deep Learning, Antonio Sclocchi, Mario Geiger, Matthieu Wyart, International Conference on Machine Learning (2023).

What can be learnt with wide convolutional neural networks?, Francesco Cagnetta, Alessandro Favero, Matthieu Wyart, International Conference on Machine Learning (2023)

How deep convolutional neural networks lose spatial information with training, Umberto Tomasini, Leonardo Petrini, Francesco Cagnetta, Matthieu Wyart, arXiv:2210.01506 (2022)

Failure and success of the spectral bias prediction for Laplace Kernel Ridge Regression: the case of low-dimensional data, Umberto Tomasini, Antonio Sclocchi, Matthieu Wyart, International Conference on Machine Learning, 21548-21583 (2022)

Learning sparse features can lead to overfitting in neural networks, Leonardo Petrini, Francesco Cagnetta, Eric Vanden-Eijnden, Matthieu Wyart, Advances in Neural Information Processing Systems 35 (2022)

Locality defeats the curse of dimensionality in convolutional teacher-student scenarios, Alessandro Favero, Francesco Cagnetta, Matthieu Wyart, Advances in Neural Information Processing Systems 34, 9456-9467 (2021)

Relative stability toward diffeomorphisms indicates performance in deep net, Leonardo Petrini, Alessandro Favero, Mario Geiger, Matthieu Wyart, Advances in Neural Information Processing Systems 34, 8727-8739 (2021)

Landscape and training regimes in deep learning, Mario Geiger, Leonardo Petrini, Matthieu Wyart, Physics Reports 924, 1-18 (2021)

Geometric compression of invariant manifolds in neural networks, Jonas Paccolat, Leonardo Petrini, Mario Geiger, Kevin Tyloo, Matthieu Wyart, Journal of Statistical Mechanics: Theory and Experiment 2021 (4), 044001 (2021)

How isotropic kernels perform on simple invariants, Jonas Paccolat, Stefano Spigler, Matthieu Wyart, Machine Learning: Science and Technology 2 (2), 025020 (2021)

Asymptotic learning curves of kernel methods: empirical data versus teacher–student paradigm, Stefano Spigler, Mario Geiger, Matthieu Wyart, Journal of Statistical Mechanics: Theory and Experiment 2020 (12), 124001 (2020)

Disentangling feature and lazy training in deep neural networks, Mario Geiger, Stefano Spigler, Arthur Jacot, Matthieu Wyart, Journal of Statistical Mechanics: Theory and Experiment 2020 (11), 113301 (2020)

Scaling description of generalization with number of parameters in deep learning, Mario Geiger, Arthur Jacot, Stefano Spigler, Franck Gabriel, Levent Sagun, Stéphane d’Ascoli, Giulio Biroli, Clément Hongler, Matthieu Wyart, Journal of Statistical Mechanics: Theory and Experiment 2020 (2), 023401

Jamming transition as a paradigm to understand the loss landscape of deep neural networks, Mario Geiger, Stefano Spigler, Stéphane d’Ascoli, Levent Sagun, Marco Baity-Jesi, Giulio Biroli, Matthieu Wyart, Physical Review E 100 (1), 012115 (2019)

A jamming transition from under-to over-parametrization affects generalization in deep learning, Stefano Spigler, Mario Geiger, Stéphane d’Ascoli, Levent Sagun, Giulio Biroli and Matthieu Wyart, Journal of Physics A: Mathematical and Theoretical 52 (47) 474001 (2019)

Comparing dynamics: Deep neural networks versus glassy systems, Marco Baity-Jesi, Levent Sagun, Mario Geiger, Stefano Spigler, Gerard Ben Arous, Chiara Cammarota, Yann LeCun, Matthieu Wyart and Giulio Biroli, Journal of Statistical Mechanics: Theory and Experiment 12, 124013 (2019)

Physics of disordered and glassy systems

Architecture of allosteric materials

Mechanics of allostery: contrasting the induced fit and population shift scenarios, Riccardo Ravasio, Solange Flatt, Le Yan, Stefano Zamuner, Carolina Brito and Matthieu Wyart, Biophysical Journal 117 (10), 1954-1962 (2019)

Direct coupling analysis of epistasis in allosteric materials, Barbara Bravi, Riccardo Ravasio, Carolina Brito and Matthieu Wyart, arXiv:1811.10480 (2018)

Principles for optimal cooperativity in allosteric materials, Le Yan, Riccardo Ravasio, Carolina Brito and Matthieu Wyart, Biophysical Journal 114 (12), 2787-2798 (2018)

Architecture and coevolution of allosteric materials, Le Yan, Riccardo Ravasio, Carolina Brito and Matthieu Wyart, PNAS 114 (10), 2526-2531 (2017)

Elasticity and mechanical stability in disordered solids

Force distribution affects vibrational properties in hard-sphere glasses, Eric DeGiuli, Edan Lerner, Carolina Brito and Matthieu Wyart, PNAS 111, 17054-17059 (2014)

Effects of coordination and pressure on sound attenuation, boson peak and elasticity in amorphous solids, Eric DeGiuli, Arnaud Laversanne-Finot, Gustavo During, Edan Lerner, Matthieu Wyart, Soft Matter, 10, 5628-5644 (2014)

Granular and suspension flows

Unified Theory of Inertial Granular Flows and Non-Brownian Suspensions, Eric DeGiuli, Gustavo During, Edan Lerner and Matthieu Wyart, Phys. Rev. E 91, 062206 (2015)

Discontinuous shear thickening without inertia in dense non-Brownian suspensions, Matthieu Wyart and Mike Cates, Phys. Rev. Lett. 112, 098302 (2014)

A unified framework for non-Brownian suspension flows and soft amorphous solids, Edan Lerner, Gustavo During, Matthieu Wyart, PNAS, 109, 4798-4803 (2012)

Glass and Rigidity transitions

Evolution of covalent networks under cooling: contrasting the rigidity window and jamming scenarios, Le Yan and Matthieu Wyart, Phys. Rev. Lett. 113, 215504 (2014)

Why glass elasticity affects the thermodynamics and fragility of super-cooled liquids, Le Yan, Gustavo During, Matthieu Wyart, PNAS, 110, 6307-6312 (2013)

Marginal Stability at Random Close Packing and other glasses

Marginal Stability in Structural, Spin, and Electron Glasses, Markus Muller and Matthieu Wyart, Annual Review of Condensed Matter Physics 6, 177-200 (2015)

Marginal Stability Constrains Force and Pair Distributions at Random Close Packing, Matthieu Wyart, Physical Review Letters,  109, 125502 (2012)

Yielding transition and elasto-plasticity

Scaling description of the yielding transition in soft amorphous solids at zero temperature, Jie Lin, Edan Lerner, Alberto Rosso and Matthieu Wyart, PNAS 111, 14382-14387 (2014)

On the density of shear transformation in amorphous solids, Jie Lin, Alaa Saade, Edan Lerner, Alberto Rosso and Matthieu Wyart, Euro. Phys. Lett., 105, 26003 (2014)