Thomas George

pic.jpg

I am a researcher at Orange Innovation in the Responsible and Impactful AI program.

My research interests include:

  • Generalization and explainability of deep neural networks
  • Weakly supervised learning
  • Causality in machine learning

I did my PhD at Mila in Québec, under the joint supervision of Pascal Vincent and Guillaume Lajoie, followed by a post-doc in Vincent Lemaire’s group. Previously, I was a multiple hats engineer at Eco-Adapt where I worked on time series from industrial sensors, in order to implement automated algorithms to make sense of these data streams. Prior to that I studied at École des Mines.

Here is my academic CV.

news

Oct 17, 2024 Our paper Mislabeled examples detection viewed as probing machine learning models: concepts, survey and extensive benchmark was accepted to TMLR. (code, video)
May 02, 2024 I joined Orange Innovation as a permanent research scientist in the Responsible and Impactful AI program.
Jul 04, 2023 I will present our paper Lazy vs hasty: linearization in deep networks impacts learning schedule based on example difficulty at Conférence sur l’Apprentissage Automatique (CAp) in Strasbourg
May 02, 2023 I am now a postdoctoral researcher at Orange Labs in Vincent Lemaire’s group.
Apr 20, 2023 I successfully defended my PhD Thesis: Deep networks training and generalization: insights from linearization (manuscript, slides) 🥳

latest posts

selected publications

  1. EKFAC
    Fast approximate natural gradient descent in a kronecker factored eigenbasis
    Thomas George, César Laurent, Xavier Bouthillier, and 2 more authors
    Advances in Neural Information Processing Systems, 2018
  2. NTKAlign
    Implicit regularization via neural feature alignment
    Aristide Baratin, Thomas George, César Laurent, and 4 more authors
    In International Conference on Artificial Intelligence and Statistics, 2021
  3. LazyHasty
    Lazy vs hasty: linearization in deep networks impacts learning schedule based on example difficulty
    Thomas George, Guillaume Lajoie, and Aristide Baratin
    Transactions on Machine Learning Research, 2022