Thomas George
Thomas George
Orange Innovation
Verified email at - Homepage
Cited by
Cited by
Fast approximate natural gradient descent in a kronecker factored eigenbasis
T George, C Laurent, X Bouthillier, N Ballas, P Vincent
Advances in Neural Information Processing Systems 31, 2018
Implicit regularization via neural feature alignment
A Baratin, T George, C Laurent, R Devon Hjelm, G Lajoie, P Vincent, ...
The 24th International Conference on Artificial Intelligence and Statistics …, 2020
Continual learning in deep networks: an analysis of the last layer
T Lesort, T George, I Rish
arXiv preprint arXiv:2106.01834, 2021
Revisiting loss modelling for unstructured pruning
C Laurent, C Ballas, T George, N Ballas, P Vincent
arXiv preprint arXiv:2006.12279, 2020
NNGeometry: easy and fast fisher information matrices and neural tangent kernels in PyTorch
T George
Lazy vs hasty: linearization in deep networks impacts learning schedule based on example difficulty
T George, G Lajoie, A Baratin
Transactions on Machine Learning Research, 2022
An evaluation of fisher approximations beyond kronecker factorization
C Laurent, T George, X Bouthillier, N Ballas, P Vincent
The Dynamics of Functional Diversity throughout Neural Network Training
L Zamparo, ME Brunet, T George, S Kharaghani, GK Dziugaite
NeurIPS 2021 Workshop BDL, 2021
A Transfer Learning Pipeline for Educational Resource Discovery with Application in Survey Generation
I Li, T George, A Fabbri, T Liao, B Chen, R Kawamura, R Zhou, V Yan, ...
Proceedings of the 18th Workshop on Innovative Use of NLP for Building …, 2023
Deep networks training and generalization: insights from linearization
T George
Factorized second order methods in neural networks
T George
The system can't perform the operation now. Try again later.
Articles 1–11