I am broadly working in the convex hull of numerical optimization, data science, and distributed methods.
Right now, I am very interested in what happens when optimization algorithms arrive near non-differentiability points, both theoretically and practically:
W. Azizian, F. Iutzeler, J. Malick, and P. Mertikopoulos: The last-iterate convergence rate of optimistic mirror descent in stochastic variational inequalities , 34th Annual Conference on Learning Theory (COLT), 2021. PDF
G. Bareilles, F. Iutzeler, J. Malick : Newton acceleration on manifolds identified by proximal-gradient methods, to appear in Mathematical Programming, 2022. PDF
F. Iutzeler, J. Malick: Nonsmoothness in Machine Learning: specific structure, proximal identification, and applications, Set-Valued and Variational Analysis, vol. 28, no. 4, pp. 661-678, 2020. PDF
Recently, I have started working on the smoothing of Wasserstein Distributionally Robust Optimization problems.
W. Azizian, F. Iutzeler, J. Malick : Regularization for Wasserstein Distributionally Robust Optimization, 2022. PDF
I am also eager to work with multi-agent systems, especially when the communications between agents are difficult (asynchronous, costly, or intermittant for example):
Y.-G. Hsieh, F. Iutzeler, J. Malick, P. Mertikopoulos : Multi-Agent Online Optimization with Delays: Asynchronicity, Adaptivity, and Optimism , Journal of Machine Learning Research, vol. 23, no. 78, pp. 1-49, 2022. PDF
D. Grishchenko, F. Iutzeler, J. Malick, M.-R. Amini: Distributed Learning with Sparse Communications by Identification , SIAM Journal on Mathematics of Data Science, vol. 3, no. 2, pp. 715-735, 2021. PDF
F. Iutzeler, J. Malick, and W. de Oliveira : Asynchronous level bundle methods, Mathematical Programming, vol. 184, pp. 319-348, 2020. PDF
Before that, I studied how randomized monotone operators could help solve practial problems in a distributed fashion:
G. Bareilles, Y. Laguel, D. Grishchenko, F. Iutzeler, J. Malick: Randomized Progressive Hedging methods for Multi-stage Stochastic Programming , Annals of Operations Research, vol. 295, no. 2, pp. 535-560, 2020. PDF Toolbox/Code
F. Iutzeler and L. Condat : Distributed Projection on the Simplex and l1 Ball via ADMM and Gossip, IEEE Signal Processing Letters, vol. 25, no. 11, pp. 1650-1654, Nov. 2018. PDF
P. Bianchi, W. Hachem, and F. Iutzeler : A Stochastic Coordinate Descent Primal-Dual Algorithm and Applications to Distributed Optimization , IEEE Transactions on Automatic Control, vol. 61, no. 10, pp. 2947-2957, Oct. 2016. PDF
Waiss Azizian (2022- at LJK, Grenoble)
Co-advised (30%) with J. Malick (LJK, Grenoble) and Panayotis Mertikopoulos (LIG, Grenoble)
Victor Mercklé (2022- at LJK, Grenoble).
Co-advised (50%) with I. Redko (Huawei, Paris, formerly HCL, St-Etienne & Aalto, Finland)
Sélim Chraibi (2019 & 2022- at LJK, Grenoble).
Co-advised (30%) with J. Malick (LJK, Grenoble)
Yu-Guan Hsieh (2019-2023 at LJK, Grenoble) now ML researcher at Apple, Paris..
Co-advised (30%) with J. Malick (LJK, Grenoble) and P. Mertikoploulos (LIG, Grenoble)
Gilles Bareilles (2019-2022 at LJK, Grenoble) now post-doc in Prague..
Co-advised (80%) with J. Malick (LJK, Grenoble)
Mathias Chastan (2019-2022 at ST MicroElectronics and LJK, Grenoble) now consultant in Data Science.
CIFRE (industrial) co-advised with J. Malick (LJK, Grenoble) and A. Lam (ST MicroElectronics, Crolles)
Dmitry Grishchenko (2017-2020 at LJK, Grenoble) now engineer at Google Munchen.
Co-advised (50%) with J. Malick (LJK, Grenoble) and M.-R. Amini (LIG, Grenoble)
Bikash Joshi (2014-2017 at LIG, Grenoble) now data-scientist in a private company.
Co-advised (50%) with M.-R. Amini (LIG, Grenoble)
Florian Vincent (Mar.-Aug. 2023 at LJK, Grenoble)
Julien Prando (Mar.-Aug. 2022 at LJK, Grenoble)
Waiss Azizian (again!!) (Mar.-Aug. 2022 at LJK, Grenoble)
Co-advised (50%) with J. Malick (LJK, Grenoble) and Panayotis Mertikopoulos (LIG, Grenoble)
Waiss Azizian (again!) (Mar.-Aug. 2021 at LJK, Grenoble)
Co-advised (50%) with J. Malick (LJK, Grenoble) and Panayotis Mertikopoulos (LIG, Grenoble)
Waiss Azizian (Mar.-Aug. 2020 at LJK, Grenoble).
Co-advised (50%) with J. Malick (LJK, Grenoble) and Panayotis Mertikopoulos (LIG, Grenoble)
Gilles Bareilles (Apr.-Sept. 2019 at LJK, Grenoble).
Yu-Guan Hsieh (Apr.-Sept. 2019 at LJK, Grenoble).
Co-advised (50%) with J. Malick (LJK, Grenoble) and Panayotis Mertikopoulos (LIG, Grenoble)
Konstantin Mishchenko (Apr.-Sept. 2017 at LJK, Grenoble).
Co-advised (50%) with J. Malick (LJK, Grenoble) and M.-R. Amini (LIG, Grenoble)
Taha Essalih (Apr.-Sept. 2017 at LEME, Paris X).
Co-advised (15%) with N. El Korso, A. Breloy (LEME, Paris X) and R. Flamary (Lagrange, Nice)
ANR - Jeune Chercheur project STROLL: Harnessing Structure in Optimization for Large-Scale Learning
PI · 145kE · 2019-2023
AI Chair – MIAI (Grenoble's AI institute) Optimization and Learning
with J. Malick (PI), P. Mertikopoulos, R. Hildenbrand (Grenoble). · funding + several PhD grants · 2019-2023
PGMO - PRMO project Distributed Optimization on Graphs with Flexible Communications
PI · 5kE · 2019-2020
CNRS INSMI and INS2I - I3A project
PI · 8kE · 2018
with M. Clausel (IECL, U. Lorraine), M.-R. Amini (LIG, Grenoble).
IDEX Grenoble Alpes - IRS project Distributed Optimization for Large-scale Learning
PI · 1 PhD funding + 3kE · 2017-2020
with J. Malick (LJK, Grenoble), M.-R. Amini (LIG, Grenoble).
IDEX Grenoble Alpes - Pedagogical Initiatives project Optimisation Distribuée pour le Big Data
approx. 30kE · 2017-2019
with J. Malick (PI), A. Iouditski, R. Hildenbrand, J. Lelong, L. Viry (LJK, Grenoble).
PGMO - PRMO project Advanced nonsmooth optimization methods for stochastic programming
12kE · 2016-2018
with J. Malick (PI) (LJK, Grenoble), W. Van Ackooij (EDF, Paris), W. de Oliveira (UERJ, Rio de Janeiro, Brésil).
GDR ISIS/GRETSI Jeunes Chercheurs project ON FIRE: Calibration des futurs grands interféromètres
co-PI · 7kE · 2016-2018
with N. El Korso (co-PI), A. Breloy (LEME, Paris X), R. Flamary (Lagrange, Nice).
Funding for my PhD thesis Distributed Estimation and Optimization over Wireless Networks by the French Defense Agency (DGA) and Institut Carnot Telecom-Eurecom