Research

Research Topics

I am interested in Learning from imbalanced data with an application to bank fraud and anomaly detection by learning an appropriate representation of the data (Metric Learning), Cost Sensitive Learning technics or by optimizing appropriate measures for this context. More recentrly, I am also interested in topics like Domain Adaptation and Semi Definite Programming.

Publications

A list of all of my publications is available on my Google Scholar or on the publication page. Below you can find a list of representative paper which illustrate my work.


A Nearest Neighbor Algorithm for Imbalanced Classification (IJAIT 2021)
[Article]
Landmark-based Ensemble Learning with Random Fourier Features and Gradient Boosting (ECML 2020)
[Article]   [Annexe]
Learning from Few Positives: a Provably Accurate Metric Learning Algorithm to Deal with Imbalanced Data (IJCAI 2020)
[Article]   [Annexe]
From Cost-Sensitive Classification to Tight F-Measure Bounds (AISTATS 2019)
[Article]   [Annexe]   [Poster]

Projects and Contracts

Project ANR DIKé : Bias, fairness and ethics of compressed NLP models (2022 - 2025) (Project Member)

Project ANR LIVES : Learning with interacting views (2019 - 2020) (Project Member, PostDoc)

PhD Supervisions

TBS - TBS - September 2022 - ... - Funded by ANR Diké - Co-supervisor, Director: Julien Velcin

Eliz Peyraud - Modeling transplant outcomes from complex data - January 2022 - ... - CIFRE Thesis with IGL (Institut Georges Lopez) - Co-supervisor, Director: Julien Jacques

Research Intern Students

TBS - Towards an Ethical Compression of Deep Learning Models (Funded by ANR DIKé)

Intership/PhD Positions

Internship offer

ERIC/Hubert Curien Laboratory (4-6 months).

During this internship, we propose that the future candidate tackle the problem of fairness in Machine Learning. More specifically, we will be looking at how the involvement of or several learners, can contribute to establishing a fairer model that performs just as well, both theoretically and practically. To do so, we will use the PAC-Bayesian framework.

Services et Administrations

COrganization and Program Committe Member

Organization Committe Member

Conférence sur l'Apprentissage Automatique (CAp) : 2021

Extraction et Gestion des Connaissances (EGC) : 2023

Multidisciplinary International Social Networks Conference (MISNC) : 2018


Program Comittee Member

Extraction et Gestion des Connaissances (EGC) : 2022

International Conference on Information Management and Big Data (SIMBig) : 2022

Reviews

Conferences

International Conference on Artificial Intelligence, Information Processing and Cloud Computing (AIIPCC) : 2019

International Conference on Artificial Intelligence and Statistics (AISTATS) : 2019

European Conference on Machine Learning (ECML) : 2017, 2018

Extraction et Gestion des Connaissances (EGC) : 2022


International Journals

Applied Stochastic Models in Business Industry (ASMB) : 2020

Expert Systems With Applications (ESWA) : 2021

Electronics (MDPI) : 2021

Mathematics (MDPI) : 2021

Pattern Recognition Letter (PRL) : 2021

Transactions on Pattern Analysis and Machine Learning (TPAMI) : 2021

Administration

Co-Responsible of Scientific Seminars - ERIC Laboratory with Mohamed-Lamine MESSAI (2022 - ...)

Contact