Associate Professor of Computer Science / Accredited to supervise research

Research

Overview

My research is centred on the notion of the epistemic grounding of artificial intelligence systems and, more specifically, on the question of how learning artificial systems acquire epistemic status with respect to the domains they are designed to model.

An AI system is epistemically grounded when its underlying theoretical assumptions are well understood, and when its behavior and latent representations can be examined, interpreted, and shown to align with the constraints of the domain it is intended to model.

In my research, this overarching question is addressed from three complementary perspectives:

  • Epistemological, through the investigation of the epistemic limits of AI systems;
  • Structural, through the study of the representations learned by AI systems;
  • Coupling-oriented, through the design of AI systems as entities coupled to their environment.

The question of epistemic grounding gives rise to a range of more specific research problems and motivates diverse empirical studies, many of which have been pursued through the work of the doctoral and Master’s students I have supervised or co-supervised.

The table below presents a selection of empirical contributions that have emerged from this research programme. The full list of doctoral and Master’s research projects is provided in the following section.

Epistemological Structural Coupling
Quantitative post-hoc explainability Learning differentiated representations Default logic
Open set recognition Feature co-adapation (Neuro-)Symbolic learning
Auto-epistemic learning Graph constrained learning Non-monotonic neuro-symbolic learning
Learning in uncertain environments    

Keywords

(in alphabetic order)

  • AI epistemology and ethics
  • (deep) machine learning
  • epistemic and aleatoric uncertainty
  • explainable AI
  • interpretability
  • neuro-symbolic AI
  • transfer learning

List of publications

Scientific supervision

PhD supervision

  • C. Schoenstein (Oct. 2025 - , co-supervision w/ O. Lecompte and Y. Nevers) / Artificial intelligence for large-scale homology relationships inference
  • A. Khudiyev (Oct. 2022 - Nov. 2025, directors: A. Jeannin-Girardon, L. Gardashova) / Scaling intelligence: a formal and practical framework for computationally unbounded AI
  • Q. Christoffel (Oct. 2021 - Nov. 2024, director: A. Deruyver) / Learning differentiated representations in deep learning Models: detection of unknown classes and interpretability
  • H. Khodji (Oct. 2020 - Nov. 2022, directors: J. Thompson & P Collet) / Deep learning and transfer learning for error detection in biological sequences
  • N. Scalzitti (Dec. 2018 - Sept. 2021, directors: J. Thompson & P. Collet) / A New trategy for genome annotation using artificial intelligence
  • R. Orhand (Oct. 2018 - Nov. 2022, directors: P. Parrend & P. Collet) / Toward autonomous and explainable artificial intelligence for uncertain environments
  • A. Ouskova (Aug. 2018 - May 2022, directors: P. Parrend & P. Collet) / Evolutionary optimization of refrigeration systems using magnetocaloric alloy

Apprenticeship training (graduate students)

  • 2016-2018
    • A. Bruyant / Robust Anonynous DAta Records
    • M. Haegelin / GPU based signal processing optimization for mass spectrometry data

Internships

  • 2026
    • N. Olejniczak / Non monotonic AI (comp. sci. grad. student)
  • 2025
    • N. Olejniczak / Non monotonic AI (comp. sci. grad. student)
    • A. Rue / Heuristic strategies for problem solving (comp. sci. undergrad. student)
  • 2024
    • T. Tass / Le quotidien d’un laboratoire de recherche (grad. student in epistemology and history of sciences)
  • 2023
    • L. Pluot / Le quotidien d’un laboratoire de recherche (grad. student in epistemology and history of sciences)
  • 2022
    • L. Wehrli / Explaining machine learning black-boxes (comp. sci. grad. student) (main advisor: S. Mark-Swecker)
  • 2021
    • Q. Christoffel / Deep Bayesian neural networks (comp. sci. grad. student)
  • 2020
    • H. Khodji / Quantification of the transferability of learned features in deep neural networks (comp. sci. grad. student)
    • Q. Christoffel / Explicability of convolutional neural networks using evolutionary algorithms (comp. sci. grad. student)
  • 2019
    • A. Oury Bah / Hybrid machine learning : combining deep learning models with evolutionary algorithms (comp. sci. grad. student)
    • A. Hutt / quantitative measure of feature transfer in deep neural networks (comp. sci. grad. student)
    • C. Mengel / ATLAS : Algebraic TopoLogy for dAta Similarity (math. grad. student)
  • 2018
    • N. Bannour / Anomaly detection in time series data (main supervisor : N. Lachiche)
  • 2017
    • N. Demeure / B cell regulatory networks modeling (comp. sci. grad. student)
    • F. Delhomme (math. grad. student) and Melchior Villa (comp. sci. grad. student) / Concentric multi-valued map (co-supervisor: Dr B. Sauvage)

Research projects (~150 hrs)

  • 2025-26
    • A. Gautheron / Hierarchical reinforcement learning (comp. sci. grad. student)
    • G. Mayer / Multi-scale learning on graphs (comp. sci. grad. student)
    • Y. Farid / Feature superposition in CNNs (comp. sci. grad. student)
  • 2024-25
    • N. Olejniczak / Non monotonic AI (comp. sci. grad. student)
    • S. Hashemi / Constrained supervised learning of latent representations (comp. sci. grad. student)
    • J. Andreolli / Transfert learning (comp. sci. grad. student) / main supervisor: R. Orhand
  • 2020-21
    • N. Mountasir / Comparing uncertainty quantifications in deep neural networks / co-supervisors : S. Marc-Zwecker and D. Bernhard (comp. sci. grad. student)
    • E. Chetouane / Semantic and artificial neural networks (comp. sci. grad. student)
  • 2019-20
    • Q. Christoffel / Explainability of convolutional neural networks using evolutionary algorithms (comp. sci. grad. student)
    • F. Nawfal / Comparative study of transfer learning methods (comp. sci. grad. student)
  • 2018-19
    • E. Kalbé / Collision detection in virtual environments: comparison of cognitive and geometric approaches (comp. sci. grad. student)
    • G. Mukunde / Deep learning for protein properties identification / co-supervisors: O. Poch & L. Moulinier (comp. sci. grad. student)
    • M. Haller / Deep learning for protein fold classification / co-supervisor: C. Mayer (comp. sci. grad. student)
    • A. Hutt / Multi-task learning, transfer learning and adaptability of deep neural networks / co-supervisor: R. Orhand (comp. sci. grad. student)
  • 2017-18
    • E. Kalbé / Non-representational approaches for agent behavior modeling (comp. sci. grad. student)
    • M. Seyer / Non-representational approaches for agent behavior modeling (comp. sci. grad. student)

Scientific dissemination

  • “Comprendre l’intelligence artificielle : usages, attentes et réalités”, conf’ voisine du Jardin des Sciences de l’université de Strasbourg, Jan. 2026
  • “Usage des IA en contexte scientifique”, 5ème Rencontre des plateformes scientifiques de recherche et de services CoRTEcS, Strasbourg, Nov. 2025
  • “Éthique de l’intelligence artificielle en santé: regard sur l’IA en contexte scientifique”, séminaire sur les techniques in silico et intelligence artificielle de l’Institut du Médicament de Strasbourg (IMS), Strasbourg, Nov. 2024
  • “Une perspective épistémologique sur l’intelligence artificielle”, séminaire à l’occasion des 10 ans du laboratoire ICube, Strasbourg, Jan. 2024
  • “L’intelligence artificielle comme système socio-techinique”, séminaire des cadres de la fondation Vincent de Paul, Strasbourg, Oct. 2023
  • “Intelligence artificielle, éthique et santé”, intervention à la conférence IA de l’ESBS “L’IA au service de la santé ; éthique et cas d’usages”, Jan. 2023
  • “Intelligence artificielle : de la technique aux enjeux sociaux”, intervention à la journée doctorale de l’université de Haute Alsace à Mulhouse, June 2022
  • “Quelle intégration des systèmes intelligents dans nos sociétés ?”, intervention à la table ronde Intelligence artificielle et démocratie, vers quels interactions et enjeux du jardin des sciences de l’Université de Strasbourg, Feb. 2020
  • “Transfer learning: review and recent advances”, workshop of the ICube research axis Data Science and Artificial Intelligence (DSAI), Nov. 2019
  • Talk at the “Journées Système 2018” in Strasbourg: “Theoretical and practical challenges of machine learning” (video, in french), Oct. 2018