Quentin BERTRAND - Home

I am a post doctoral researcher at Mila working with Gauthier Gidel and Simon Lacoste-Julien. On July 1st 2024 I will join Inria as a Research Scientist! I work on automatic differentiation, games, and representation learning. Prior to this position, I did my Ph. D. at Inria Paris-Saclay (in the Parietal Team) under the supervision of Joseph Salmon and Alexandre Gramfort. I worked on the optimization and statistical aspects of high dimensional sparse linear regression applied to brain signals reconstruction. In particular, I developed python packages for fast computation and automatic hyperparameter selection of sparse linear models. Here is a short resume and my list of publications.

Contact

News

• 01-07-2023 On July 1st 2024 I will join Inria as a Research Scientist!

• 05-12-2023 I will present our paper On the Limitations of Elo: Real-World Games are Transitive, not Additive at the Berkeley Multi-Agent Reinforcement Learning Seminar

• Our paper Synergies between Disentanglement and Sparsity: Generalization and Identifiability in Multi-Task Learning has been accepted at ICML 2023, see you in Hawaii!

• Our paper On the Limitations of Elo: Real-World Games are Transitive, not Additive has been accepted to AISTATS 2023, see you in Spain!

• I just presented our paper Synergies between Disentanglement and Sparsity: a Multi-task Learning Perspective at the Canadian Mathematical Society Winter Workshop

• I just presented our two papers Beyond L1: Faster and Better Sparse Models with skglm and The Curse of Unrolling: Rate of Differentiating Through Optimization at NeurIPS 2022

• We just released our paper on sparsity for disentanglement Synergies between Disentanglement and Sparsity: a Multi-task Learning Perspective!

• I was awarded the top reviewer award at NeurIPS 2022!

• Our papers Beyond L1: Faster and Better Sparse Models with skglm and The Curse of Unrolling: Rate of Differentiating Through Optimization have been accepted to NeurIPS 2022!

• Our paper Implicit differentiation for fast hyperparameter selection in non-smooth convex learning was accepted to JMLR!

Past events

• The code of our paper Beyond L1: Faster and Better Sparse Models with skglm was integrated in the scikit-contrib repository. Thanks a lot to all the contributors!

• Our paper Electromagnetic neural source imaging under sparsity constraints with SURE-based hyperparameter tuning was accepted at the “medical imaging meets NeurIPS2021” workshop!

• Since November 2021 I am a post doctoral researcher at Mila working with Gauthier Gidel and Simon Lacoste-Julien

• I was awarded the NeurIPS 2021 outstanding reviewer award (top 8%)

• I defended my Ph. D. thesis on Tuesday 28th September! Here is the manuscript, here are the slides, and here is the video of the defense. Thanks again to all my collaborators

• February 8: I presented our work on implicit differentiation at Google Brain, Montréal

• January 23: our paper Anderson acceleration of coordinate descent was accepted to AISTATS 2021!

• January 4-9: I gave a tutorial on optimization for machine learning at the Data Science Summer School of École polytechnique

• August 26-28: I presented our paper on Support recovery and sup-norm convergence rates for sparse pivotal estimation at AISTATS 2020 (virtually)

• July 12-19: I presented our work on Implicit differentiation of Lasso-type models for hyperparameter optimization at ICML 2020 (virtually)

• June 28 - July 10, 2020: I attended the Machine Learning Summer School (MLSS) (virtually)

• January 6, 2020: our paper Support recovery and sup-norm convergence rates for sparse pivotal estimation was accepted to AISTATS 2020 in Palermo, Italy

• December 8-15, 2019: I presented our paper Handling correlated and repeated measurements with the smoothed multivariate square-root Lasso at NeurIPS 2019 in Vancouver, Canada

• October 21-25, 2019: I was awarded a grant from GDR IA to visit Samuel Vaiter in Dijon, France