A jamming transition from under- to over-parametrization affects generalization in deep learning - Archive ouverte HAL
Article Dans Une Revue Journal of Physics A: Mathematical and Theoretical Année : 2019

A jamming transition from under- to over-parametrization affects generalization in deep learning

Dates et versions

hal-02569109 , version 1 (11-05-2020)

Identifiants

Citer

S Spigler, M Geiger, Stéphane d'Ascoli, L Sagun, Giulio Biroli, et al.. A jamming transition from under- to over-parametrization affects generalization in deep learning. Journal of Physics A: Mathematical and Theoretical, 2019, 52 (47), pp.474001. ⟨10.1088/1751-8121/ab4c8b⟩. ⟨hal-02569109⟩
52 Consultations
0 Téléchargements

Altmetric

Partager

More