Convit: Improving vision transformers with soft convolutional inductive biases S d’Ascoli, H Touvron, ML Leavitt, AS Morcos, G Biroli, L Sagun International Conference on Machine Learning, 2286-2296, 2021 | 721 | 2021 |
Scaling description of generalization with number of parameters in deep learning M Geiger, A Jacot, S Spigler, F Gabriel, L Sagun, S d’Ascoli, G Biroli, ... Journal of Statistical Mechanics: Theory and Experiment 2020 (2), 023401, 2020 | 206 | 2020 |
Jamming transition as a paradigm to understand the loss landscape of deep neural networks M Geiger, S Spigler, S d'Ascoli, L Sagun, M Baity-Jesi, G Biroli, M Wyart Physical Review E 100 (1), 012115, 2019 | 160 | 2019 |
Double trouble in double descent: Bias and variance (s) in the lazy regime S d’Ascoli, M Refinetti, G Biroli, F Krzakala International Conference on Machine Learning, 2280-2290, 2020 | 143 | 2020 |
A jamming transition from under-to over-parametrization affects generalization in deep learning S Spigler, M Geiger, S d’Ascoli, L Sagun, G Biroli, M Wyart Journal of Physics A: Mathematical and Theoretical 52 (47), 474001, 2019 | 130 | 2019 |
Electromagnetic Emission from Supermassive Binary Black Holes Approaching Merger S d'Ascoli, SC Noble, DB Bowen, M Campanelli, JH Krolik, V Mewes The Astrophysical Journal 865 (2), 2018 | 102 | 2018 |
End-to-end symbolic regression with transformers PA Kamienny, S d'Ascoli, G Lample, F Charton Advances in Neural Information Processing Systems, 8754-8765, 2022 | 97 | 2022 |
Triple descent and the two kinds of overfitting: Where & why do they appear? S d'Ascoli, L Sagun, G Biroli Advances in Neural Information Processing Systems 33, 3058-3069, 2020 | 88 | 2020 |
A jamming transition from under-to over-parametrization affects loss landscape and generalization S Spigler, M Geiger, S d'Ascoli, L Sagun, G Biroli, M Wyart NeurIPS 2018 Workshop « Science of Deep Leaning Meets Engineering », 2019 | 67 | 2019 |
Deep symbolic regression for recurrence prediction S d’Ascoli, PA Kamienny, G Lample, F Charton International Conference on Machine Learning, 4520-4536, 2022 | 54* | 2022 |
Finding the Needle in the Haystack with Convolutions: on the benefits of architectural bias S d'Ascoli, L Sagun, G Biroli, J Bruna Advances in Neural Information Processing Systems, 9334-9345, 2019 | 39 | 2019 |
Align, then memorise: the dynamics of learning with feedback alignment M Refinetti, S d’Ascoli, R Ohana, S Goldt International Conference on Machine Learning, 8925-8935, 2021 | 23 | 2021 |
On the interplay between data structure and loss function in classification problems S d'Ascoli, M Gabrié, L Sagun, G Biroli Advances in Neural Information Processing Systems 34, 2021 | 17* | 2021 |
Length generalization in arithmetic transformers S Jelassi, S d'Ascoli, C Domingo-Enrich, Y Wu, Y Li, F Charton arXiv preprint arXiv:2306.15400, 2023 | 15 | 2023 |
The dynamics of learning with feedback alignment M Refinetti, S d’Ascoli, R Ohana, S Goldt Journal of Physics A, 2020 | 14 | 2020 |
Transformed CNNs: recasting pre-trained convolutional layers with self-attention S d'Ascoli, L Sagun, G Biroli, A Morcos arXiv preprint arXiv:2106.05795, 2021 | 5 | 2021 |
Conditioned Text Generation with Transfer for Closed-Domain Dialogue Systems S d’Ascoli, A Coucke, F Caltagirone, A Caulier, M Lelarge International Conference on Statistical Language and Speech Processing, 23-34, 2020 | 5* | 2020 |
Optimal learning rate schedules in high-dimensional non-convex optimization problems S d'Ascoli, M Refinetti, G Biroli arXiv preprint arXiv:2202.04509, 2022 | 4 | 2022 |
Comprendre la révolution de l'intelligence artificielle S d'Ascoli First, 2020 | 4 | 2020 |
Odeformer: Symbolic regression of dynamical systems with transformers S d'Ascoli, S Becker, A Mathis, P Schwaller, N Kilbertus arXiv preprint arXiv:2310.05573, 2023 | 3 | 2023 |