About me

CV (in french)

Since Oct. 2014, I am an associate professor (maître de conférences, HDR) at University Jean Monnet of Saint-Etienne (France). I am a member of the Hubert Curien Laboratory within the Machine Learning team. Since 2026, I have been in charge of the Foundation and Theory of Machine Learning subgroup. I have been also responsible of the 2nd year of BSc in Computer Science since July 2015 (and of the 1st year during the 2024–2025 academic year). From Oct. 2013 to Sept. 2014, I was a postdoctoral researcher at ISTA (Institute of Science and Technology Austria) in the Christoph Lampert group (Computer Vision and Machine Learning).

I obtained my Ph.D. in computer science (machine learning) at the Laboratoire d’Informatique Fondamentale (now LIS) of Marseille in the Qarma team, under the direction of Amaury Habrard and Stéphane Ayache. The main objective of my thesis was to study the learning of majority vote for supervised classification and domain adaptation. This work was supported by the ANR project VideoSense.

Research Interests

• Statistical Machine Learning
• PAC-Bayesian bounds and algorithms (see our tutorial on PAC-Bayes with P. Viallard, given at COLT 2025)
• Transfer learning, domain adaptation
• Supervised learning
• Robustness and adversarial learning
• Fairness and biais in machine learning
• Representation learning
• Similarity and metric learning
• Multiview learning
• Imbalanced data
• Multi-armed bandits

Research Projects

• ANR project Famous (local coordinator) : Fair multimodal Learning
• Projet inter-laboratoire FIL (Fédération Informatique de Lyon) (with G. Metlzer): PAC-Bayesian Fair Learning
• Projet inter-laboratoire FIL (with G. Metlzer) : Random Fourrier Features for PAC\-Bayesian Domain Adaptation
• ANR project TAUDoS (member): Theory and Algorithms for the Understanding of Deep learning On Sequential data
• ANR project APRIORI (coordinator): A Pac-bayesian RepresentatIOn leaRnIng perspective
• INS2I-CNRS project JCJC 2018 “PaRaFF” (with P. Germain): PAC-Bayesian Random Fourier Features
• ANR project LIVES (member) : Learning with Interacting ViEwS
• My postdoc was funded by the ERC grant no308036: Life-Long Learning for Visual Scene Understanding
• I was a member of the PASCAL2 Network of excellence
• My Ph.D. was funded by VideoSense ANR project: Rich concepts recognition in multilingual videos, throw pivot languages
• I regularly participated in Lampada ANR Project: Learning Algorithms, Models and sPArse representations for structured DAta

Supervision

Current Ph.D. Student

• Hind Atbir — Learning fair and robust kernel-based models with generalization guarantees (from Oct. 2024 to … , co-supervised with Farah Cherfaoui and Paul Viallard)

• Julien Bastian — Multiview Fair Learning - From Theory to Algorithms (from Oct. 2024 to … , co-supervised with Christine Largeron and Guillaume Metzler)

Former Ph.D. Students

• Paul Viallard — Beyond PAC-Bayesian Bounds: From Disintegration to Novel Bounds (from Sept. 2019 to Dec. 2022, co-supervised with Amaury Habrard and Pascal Germain) — Now IFSP Inria Rennes

• Léo Gautheron — Learning Tailored Data Representations from Few Labeled Examples (from Oct. 2017 to Dec. 2020, co-supervised with Amaury Habrard and Marc Sebban) — Now Data Science Engineer at Synapse Défense

• Anil Goyal — Learning a Multiview Weighted Majority Vote Classifier: Using PAC-Bayesian Theory and Boosting (from Nov. 2015 to Oct. 2018, co-supervised with Massih-Reza Amini) — Now Principal Data Scientist at Mastercard

Former Post-doc

• Marie-Ange Lebre — Deep Learning for detection and classification of microorganism (from Sept. 2022 to Sept. 2024, co-supervised with Amaury Habrard and Rémi Emonet)

Interns

• David Nguyen Phung — PAC-Bayes and Multi-Armed Bandits (Apr-July 2026, co-supervised with P. Viallard)
• Circée Chalayer — (Feb-July 2026, co-supervised with R. Emonet and Q. Bertrand)
• Solal Peiffer-Smadja — PAC-Bayes and distributionally robust optimization (Feb-July 2026, co-supervised with H. Atbir, F. Cherfaoui and P. Viallard)
• Baptiste Mathevon — PAC-Bayes and Multi-Armed Bandits (Apr-July 2025, co-supervised with P. Viallard)
• Julien Bastian — Fairness and domain generalization (Mar-Aug 2024, co-supervised with G. Metzler)
• Hind Atbir — PAC-Bayesian Fair Learning (Feb-July 2024, co-supervised with F. Cherfaoui, G. Metzler and P. Viallard)
• Mickaël Gault — Learning fair kernel classifier under constraints (Mar-July 2024, co-supervised with G. Metzler)
• Julien Bastian — Random Fourier Features, PAC-Bayes and Domain Adaptation (Mar-July 2023, co-supervised with G. Metzler)
• Alexiane Fraisse — Random Fourier Features and Domain Adaptation (Apr-July 2022, co-supervised with G. Metzler and P. Viallard)
• Luiza Dzhidzhavadze — A Multiclass C-Bound-Based Algorithm (Apr-June 2021, co-supervised with P. Viallard)
• Himanshu Pandey — A Multiclass C-Bound-Based Algorithm (Apr-June 2021, co-supervised with P. Viallard)
• Paul Viallard — Deep Learning and PAC-Bayes (Feb-June 2019, co-supervised with A. Habrard and R. Emonet)
• Omar El-Sabrout — Active Learning for PAC-Bayesian Domain Adaptation (Mid-April Mid-July 2018)
• Loujain Liekah — Experts Combination (Apr-June 2018, co-supervised with M. Soare)
• Luc Giffon — Efficient anomaly detection in data stream (Feb-June 2017, co-supervised with A. Bonnefoy and T. Peel)
• Arunava Maulik — (Apr-June 2017, co-supervised with A. Habrard and M. Soare)
• Prem Prakash — (Apr-June 2017, co-supervised with A. Habrard)
• Léo Gautheron — Improving the bibliometry platform Labmetry (Apr-June 2016, co-supervised with M. Sebban)
• Benjamin Sabot — Empirical study of the C-bound as stopping criterion for neural networks (Apr-June 2016, co-supervised with A. Habrard, P. Germain and D. Fourure)
• Soroush Seifi — A PAC-Bayesian Multiview Study (Apr-June 2016, co-supervised with A. Habrard and A. Goyal)

PhD Jurys

• 2025 : Thibaud Leteno, St-Etienne (invited)
• 2022 : Tahar Allouch, Paris (external examinator), Luxin Zhang, Lille (external examinator), Paul Viallard (co-supervisor)
• 2020 : Léo Gautheron, St-Etienne (co-supervisor)
• 2018 : Anil Goyal, St-Etienne/Grenoble (co-supervisor)

Current Teaching

Depuis Juillet 2015, je suis la responsable de la 2ème année de la licence d’informatique. Depuis Juillet 2024, je suis également responsable de la 1ère année de la licence d’informatique

Since July 2015, I am in charge of the 2nd year of BSc in Computer Science. Since July 2024, I am also in charge of the 1st year of BSc in Computer Science.

• L1 Info - Programmation fonctionnelle - Ocaml
• L2 Info - Systèmes d’exploitation
• L2 Info - Programmation impérative - C (classique et spécifique alternants)

Administration activities

• 2024-2025 : Responsible of the 1st year of BSc in Computer Science
• 2021-2022: Member of CNU 27 (French Council of Universities for Computer Science)
• 2019: Board Member of the machine learning and artificial intelligence (MALIA) group of the French association on Statistics (SFdS)
• 2018: “Founding member” of the MALIA group
• 2017-2020: Vice-President (and founding member) of the French Association on Machine Learning (SSFAM)
• 2015-now: Responsible of the 2nd year of BSc in Computer Science
• 2015-now: Member of the board of the Hubert Curien Laboratory

Member of Organization Committees

• 2022: Publicity chair of ECML-PKDD 2022
• 2019: Demonstration co-chair at ECML-PKDD 2019
• Oct. 2015: Member of the local team of IDA 2015 (International Symposium on Intelligent Data Analysis)
• Sept. 14: Organization of the international workshop on LEarning with Multiple views: Applications to computer vision and multimedia (in conjunction with ECML-PKDD 2014), with S. Ayache, M. Cord, and F.-X. Dupé
• May 14: Organization of the annual conference of the Austrian Association for Pattern Recognition (OAGM 2014), with V. Kolomogorov, C.H. Lampert, and R. Takhanov.

Member of Program Committees

• 2025: CAp, NeurIPS
• 2022: ECML-PKDD (as area chair), ICML, CAp
• 2021: ICML, CAp
• 2020: ICML, IDA, CAp
• 2019: ICML, ECML-PKDD, CAp
• 2018: ICML, CAp
• 2017: AISTATS, ICML, NIPS
• 2016: ICML, NIPS, BeyondLabeler, CAp, TASK-CV
• 2015: ICML, TASK-CV
• 2014: ICPRAM, ECAI, ECCV, TASK-CV
• 2013: CAp
• Journals: JMLR, TPAMI, Pattern Recogn. Lett.

Invited talks / Seminaries / Tutorials

• June 25: Conference on Learning Theory (COLT 2025), Lyon, France: How to Make Use of Learning Theory to Learn Efficient ML Models: From PAC-Bayesian Generalization Bounds to (Self-Bounding) Learning Algorithms (Tutorial with Paul Viallard) [Ressources]
• June 19: Journées de Statistique 2019, Nancy, France: When PAC\-Bayesian Majority Vote meets Domain Adaptation
• June 18: “Les Universitaires retournent à l’École”, Lycée Étienne Mimard, Saint-Etienne, France: Apprentissage Automatique et Adaptation de Domaine
• Feb. 18: MODAL Seminars, INRIA Lille, France: When PAC\-Bayesian Majority Vote Meets Transfer Learning [SLIDES]
• Jan. 18: Visit of “Université pour tous” at LaHC, Univ. of Saint-Etienne, France: Presentation of the Data Intelligence Groupand What is Domain Adaptation?
• Jan. 16: Visit of students at LaHC, Univ. of Saint-Etienne, France: What is Domain Adaptation? A popularization
• Jan. 16: ANR project LIVES workshop, Aix-Marseille Univ., France: PAC-Bayesian Majority Vote & Domain Adaptation
• Feb. 15: Machine Learning Seminars LaHC, Univ. of Saint-Étienne, France: Multilabel Structured Output Learning with A Random Sample of Spanning Trees
• May 14: AMA Seminars, LIG, Grenoble, France: When PAC-Bayes meets Domain Adaptation
• Feb. 14: Machine Learning Seminars LaHC, Univ. of Saint-Étienne, France: Domain Adaptation of Majority Votes via Perturbed Variation-based Label transfer
• Dec. 13: Signal Processing - Machine Learning Seminars LATP/LIF, Aix-Marseille Univ., France: Domain Adaptation of Majority Votes via Perturbed Variation-based Label transfer
• Lampada Workshop, Porquerolles, France: A PAC-Bayesian Approach for Domain Adaptation
• April 13: IST Austria, Klosterneuburg, Austria: Combining Similarities or Classifiers for Domain Adaptation
• March 13: Xerox Research Center Europe, Grenoble, France: Combining Similarities or Classifiers for Domain Adaptation
• Nov. 12: Signal Processing - Machine Learning Seminars LATP/LIF, Aix-Marseille Univ., France: A Well-founded PAC-Bayesian Majority Vote applied to the Nearest Neighbor Rule
• Aug. 12: GRAAL Seminars, Univ. Laval, Québec, Canada: Unsupervised and Semi-supervised Domain Adaptation with Good Similarity Functions
• June 12: Lampada Workshop, Lille, France: PAC-Bayes Bound and Multiclass Classification
• April 12: VideoSense Meeting, Grenoble, France: From PAC-Bayesian MinCq to Late Classifier Fusion
• March 12: HIIT (Helsinki Institute for Information Technology) Seminars, Espoo, Finland: A General Framework for Domain Adaptation in a Good Similarity-Based Projection Space
• Sept. 11: Signal Processing - Machine Learning Seminars LATP/LIF, Aix-Marseille Univ., France: Sparse Domain Adaptation in Projection Space based on Good Similarity Function
• June 11: Lampada Workshop, Saint-Victor sur Loire, France: Domain Adaptation with Good Similarity Functions
• Oct. 10: VideoSense Meeting, Sophia-Antipolis, France: Domain Adaptation Algorithm for Learning Classifier

Spare Time Activities

• I am a 3rd star black belt in Manchuria Kung Fu, I am also an instructor of Manchuria Kung Fu. I have been trained by Master Mathieu Derosière.
• I have been playing video games since I was 4 or 5 years old. I started to play on an Amstrad CPC with a cassette tape deck!