A statistical mechanics framework for Bayesian deep neural networks beyond the infinite-width limit R Pacelli, S Ariosto, M Pastore, F Ginelli, M Gherardi, P Rotondo Nature Machine Intelligence 5 (12), 1497-1507, 2023 | 32* | 2023 |
Learning through atypical phase transitions in overparameterized neural networks C Baldassi, C Lauditi, EM Malatesta, R Pacelli, G Perugini, R Zecchina Physical Review E 106 (1), 014116, 2022 | 27 | 2022 |
Local Kernel Renormalization as a mechanism for feature learning in overparametrized Convolutional Neural Networks R Aiudi, R Pacelli, A Vezzani, R Burioni, P Rotondo arXiv preprint arXiv:2307.11807, 2023 | 7 | 2023 |
Universal mean-field upper bound for the generalization gap of deep neural networks S Ariosto, R Pacelli, F Ginelli, M Gherardi, P Rotondo Physical Review E 105 (6), 064309, 2022 | 3 | 2022 |
Predictive Power of a Bayesian Effective Action for Fully Connected One Hidden Layer Neural Networks in the Proportional Limit P Baglioni, R Pacelli, R Aiudi, F Di Renzo, A Vezzani, R Burioni, ... Physical Review Letters 133 (2), 027301, 2024 | 2 | 2024 |
Statistical mechanics of transfer learning in fully-connected networks in the proportional limit A Ingrosso, R Pacelli, P Rotondo, F Gerace arXiv preprint arXiv:2407.07168, 2024 | | 2024 |
A data-agnostic statistical mechanics approach for studying deep neural networks beyond the infinite-width limit R Pacelli Politecnico di Torino, 2024 | | 2024 |