Since 2022, I am a tenure-track assistant professor (maître de conférences) at the Laboratoire des Signaux et Systèmes at CentraleSupélec. Prior to that, I was a post-doc in the team of Tim van Erven at the Korteweg-de Vries Institute for Mathematics, of the University of Amsterdam. I defended my PhD in 2020 at the LMO of the Université Paris-Saclay under the supervision of Gilles Stoltz and Pascal Massart.
My research interests lie broadly in the mathematics of online decision-making, focusing on adaptivity in bandits and online learning.
Office : A5.02 in the
Bâtiment Bréguet
Mail : hedi.hadiji@l2s.centralesupelec.fr or hedi.hadiji@gmail.com
Diversity-preserving K-armed bandits, revisited, with Sébastien Gerchinovitz, Jean-Michel Loubes, and Gilles Stoltz, 2024, (Updated and improved version of our 2020 preprint) [HAL]
Towards Characterizing the First-order Query Complexity of Learning (Approximate) Nash Equilibria in Zero-sum Matrix Games, with Wouter Koolen, Sarah Sachs, Tim van Erven, 2023, accepted to Neurips '23 [arxiv]
Accelerated Rates between Stochastic and Adversarial Online Convex Optimization, with Tim van Erven, Cristóbal Guzmán, and Sarah Sachs, 2023, (Extended version of our 2022 Neurips paper) [arxiv]
Scale-free Unconstrained Online Learning for Curved Losses, with Tim van Erven and Jack Mayo, 2022, Conference on Learning Theory (COLT) 2022 [arxiv] [slides]
Between Stochastic and Adversarial Online Convex Optimization: Improved Regret Bounds via Smoothness, with Tim van Erven, Cristóbal Guzmán, and Sarah Sachs, 2022, Advances in Neural Information Processing Systems (NeurIPS), 2022 [arxiv] [slides]
Distributed Online Learning for Joint Regret with Communication Constraints, with Tim van Erven and Dirk van der Hoeven, Algorithmic Learning Theory (ALT) 2022 [arxiv] [slides]
Adaptation to the Range in K-Armed Bandits, with Gilles Stoltz, Journal of Machine Learning Research, 2023 [HAL]
Polynomial Cost of Adaptation for X-armed Bandits, Advances in Neural Information Processing Systems (NeurIPS), 2019 [arxiv] [ poster]
KL-UCB-switch: Optimal Regret Bounds for Stochastic Bandits from both a Distribution-dependent and a Distribution-free Viewpoints, with Aurélien Garivier, Pierre Ménard and Gilles Stoltz, Journal of Machine Learning Research, 2022 [arxiv]
DATAIA - ILLS workshop, CentraleSupélec, May 2023
Control Theory and Inverse Problems conference, Monastir, May 2023
LTCI S2A Seminar, Télécom Paris, July 2022
KdVI Statistics seminar, University of Amsterdam, March 2022
Séminaire de Mathématiques appliquées du Laboratoire Jean Leray, Université de Nantes February 2022
Weekly Machine Learning seminar, CWI, May 2021
Thematic Statistics Seminar in Machine Learning, Leiden University, February 2020
Seminar of the Institut für Mathematische Stochastik (IMST), Otto-von-Guericke-Universität, October 2019
Workshop of the Celeste INRIA team, INRIA September 2019
Séminaire des doctorants du Laboratoire de Mathématiques d'Orsay, Université Paris-Saclay, June 2019
50èmes Journées de Statistique, EDF Lab Paris Saclay, May 2018
My PhD thesis, and the slides from my defense.
Reviewing services: COLT('23, '22, '21), UAI '21, Neurips'19 (top 400 reviewer), JMLR, Mathematics of Operations Research, SIMODS, Optimization Letters
I enjoy implementing the algorithms I design: my personal python bandit laboratory for numerical experiments.
2023-2024:
2022-2023:
2020-2022: TA at University of Amsterdam
2019-2020: TA at Université Paris-Saclay (then Paris-Sud)
2017-2019: TA at the IUT de Sceaux
I enjoy bouldering and I'd like to be better at chess. I read a lot of comics/graphic novels, and sometimes real books. When I feel free, I like to go to the movies.
My first name is Hédi and my last name is Hadiji. In Arabic my first name is written هادي. In English (and in French) pronounce my name like "Eddy".