| Entropy-sgd: Biasing gradient descent into wide valleys P Chaudhari, A Choromanska, S Soatto, Y LeCun, C Baldassi, C Borgs, ... arXiv preprint arXiv:1611.01838, 2016 | 642 | 2016 |
| 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 | 425 | 2021 |
| Searchqa: A new q&a dataset augmented with context from a search engine M Dunn, L Sagun, M Higgins, VU Guney, V Cirik, K Cho arXiv preprint arXiv:1704.05179, 2017 | 375 | 2017 |
| Empirical analysis of the hessian of over-parametrized neural networks L Sagun, U Evci, VU Guney, Y Dauphin, L Bottou arXiv preprint arXiv:1706.04454, 2017 | 278 | 2017 |
| 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 | 177 | 2020 |
| Eigenvalues of the hessian in deep learning: Singularity and beyond L Sagun, L Bottou, Y LeCun arXiv preprint arXiv:1611.07476, 2016 | 176* | 2016 |
| A tail-index analysis of stochastic gradient noise in deep neural networks U Simsekli, L Sagun, M Gurbuzbalaban International Conference on Machine Learning, 5827-5837, 2019 | 167 | 2019 |
| 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 | 156* | 2019 |
| 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 | 134 | 2019 |
| Comparing dynamics: Deep neural networks versus glassy systems M Baity-Jesi, L Sagun, M Geiger, S Spigler, GB Arous, C Cammarota, ... International Conference on Machine Learning, 314-323, 2018 | 105 | 2018 |
| Energy landscapes for machine learning AJ Ballard, R Das, S Martiniani, D Mehta, L Sagun, JD Stevenson, ... Physical Chemistry Chemical Physics 19 (20), 12585-12603, 2017 | 103 | 2017 |
| 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 | 66 | 2020 |
| Explorations on high dimensional landscapes L Sagun, VU Guney, GB Arous, Y LeCun arXiv preprint arXiv:1412.6615, 2014 | 64 | 2014 |
| Vision models are more robust and fair when pretrained on uncurated images without supervision P Goyal, Q Duval, I Seessel, M Caron, I Misra, L Sagun, A Joulin, ... arXiv preprint arXiv:2202.08360, 2022 | 53 | 2022 |
| Early Predictability of Asylum Court Decisions M Dunn, H Sirin, L Sagun, D Chen | 36* | 2017 |
| 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 32, 2019 | 30 | 2019 |
| On the heavy-tailed theory of stochastic gradient descent for deep neural networks U Şimşekli, M Gürbüzbalaban, TH Nguyen, G Richard, L Sagun arXiv preprint arXiv:1912.00018, 2019 | 20 | 2019 |
| 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, 8506-8517, 2021 | 14* | 2021 |
| Easing non-convex optimization with neural networks D Lopez-Paz, L Sagun | 12 | 2018 |
| Universal halting times in optimization and machine learning L Sagun, T Trogdon, Y LeCun arXiv preprint arXiv:1511.06444, 2015 | 12* | 2015 |
Giulio BiroliProfessor of Theoretical Physics, ENS ParisVerified email at cea.fr
