An inertial Newton algorithm for deep learning (2021)

C. Castera, J. Bolte, C. Févotte, E. Pauwels
Journal of Machine Learning Research

JMLR | Hal | Code

Second-order step-size tuning of SGD for non-convex optimization (2022)

C. Castera, J. Bolte, C. Févotte, E. Pauwels
Neural Processing Letters

NePL | arXiv

Inertial Newton algorithms avoiding strict saddle points (2021)

C. Castera
Preprint

arXiv

Source code for the INNA algorithm

Pytorch version | Keras version | Tensorflow version

Some neural networks implemented in Keras

Networks for image classification