Research

Papers

• 2023

  • J. Larsson, Q. Klopfenstein, M. Massias, J. Wallin, Coordinate descent for SLOPE, AISTATS 2023.

  • B. Moufad, P.-A. Bannier, Q. Bertrand, Q. Klopfenstein, M. Massias, skglm: improving scikit-learn for regularized Generalized Linear Models, submitted to JMLR OSS.

• 2022

  • T. Moreau, M. Massias, A. Gramfort and others, Benchopt: Reproducible, efficient and collaborative optimization benchmarks, NeuRIPS 2022.

  • Q. Bertrand, Q. Klopfenstein, P.-A. Bannier, G. Gidel, M. Massias, Beyond L1: Faster and better sparse models with skglm, NeuRIPS 2022.

  • B. Muzellec, K. Sato, M. Massias, T. Suzuki, Dimension-free convergence rates for gradient Langevin dynamics in RKHS, COLT 2022.

  • C. Molinari, M. Massias, L. Rosasco, S. Villa, Iterative regularization for low-complexity regularizers, submitted.

• 2021

  • Q. Bertrand, Q. Klopfenstein, M. Massias, M. Blondel, S. Vaiter, A. Gramfort, J. Salmon, Implicit differentiation for fast hyperparameter selection in non-smooth convex learning, JMLR. code

  • Q. Bertrand, M. Massias, Anderson acceleration of coordinate descent, AISTATS 2021. code

  • C. Molinari, M. Massias, L. Rosasco, S. Villa, Iterative regularization for convex regularizers, AISTATS 2021. code

• 2020

  • M. Massias*, Q. Bertrand*, A. Gramfort, J. Salmon, Support recovery and sup-norm convergence rates for sparse pivotal estimation, AISTATS 2020.

  • M. Massias, S. Vaiter, A. Gramfort, J. Salmon, Dual extrapolation for sparse Generalized Linear Models, JMLR. celer library

• 2019

  • P. Ablin, T. Moreau, M. Massias, A. Gramfort, Learning step sizes for unfolded sparse coding, NeurIPS 2019.

  • Q. Bertrand*, M. Massias*, A. Gramfort, J. Salmon, Handling correlated and repeated measurements with the smoothed multivariate square-root Lasso, NeurIPS 2019. code

• 2018

  • M. Massias, A. Gramfort, J. Salmon, Celer: a fast solver for the Lasso with dual extrapolation, ICML 2018. slides code doc

  • M. Massias, O. Fercoq, A. Gramfort, J. Salmon, Generalized concomitant multi-task Lasso for sparse multimodal regression, AISTATS 2018. code

• 2017

  • M. Massias, A. Gramfort, J. Salmon, From safe screening rules to working sets for faster Lasso-type solvers, OPTML workshop at NIPS 2017. deprecated code

  • M. Massias, J. Salmon, A. Gramfort, Gap safe screening rules for faster complex-valued multi-task group Lasso, SPARS, Lisbon, 2017.

PhD. Thesis

• M. Massias, Sparse high dimension linear regression in the presence of heteroscedastic noise: application to magnetoelectric source imaging. Defended on 04122019. manuscript slides

Relevant slides

• Dual extrapolation, 09032020, Optimization for Machine Learning workshop at CIRM, Luminy.

• LCSL group meeting, 23012020, Università di Genova.

• PhD Defense, 04122019, Inria.

• Dual extrapolation for sparse Generalized Linear Models, 09072019, Applied Inverse Problems conference, Grenoble.

• Celer: fast solver for the Lasso with dual extrapolation, 11052018, Statistics Seminar at University of Washington.

• Generalized concomitant multi-task Lasso for sparse multimodal regression, 20102017, Journées GDR MOA MIA (Bordeaux). Also presented at PGMO days 2017, Saclay.

• Résolution rapide de problèmes de type Lasso: des règles de safe screening aux working sets (in French), 05092017, GRETSI. Also presented at JDSE 2017 (best presentation award).

• Faster solvers for sparse multi-task problems, 20032017, IAP (Paris). Also presented at CMStats 2017, London.