Short Bio
I am an Associate Professor at Universite Lumiere Lyon 2 and a member of LIRIS research laboratory. Before that, I was a postdoctoral researcher at Mines Saint-Etienne from 2024 to 2025. I received my PhD in Computer Science from Universite Cote d'Azur and hold engineering degrees from ENSAI Rennes and ENSAE Dakar, as well as a Master's degree in Mathematics and Applications, with a specialization in Pure Mathematics. My research focuses on deep learning, anomaly detection, computational cost, energy efficiency, and sustainable AI.
Research
I develop frugal / Green AI methods to reduce the carbon and energy footprint of machine learning models. I study optimization and compression techniques tailored to resource-constrained settings. In parallel, I work on anomaly and drift detection for time-series data, with applications in health, industry, and the environment.
Topics
- Efficient deep learning: training/inference cost, energy-aware evaluation, scaling laws
- Model optimization & compression for constrained deployments
- Anomaly and drift detection in time series (predictive maintenance, health monitoring)
Collaboration interests
I am particularly interested in collaborations on Green / frugal AI (compression, optimization, deployment), and time-series anomaly detection (health, industry, environment).
Generalization and Scaling Laws for Mixture-of-Experts Transformers
This paper studies how sparse Mixture-of-Experts Transformers generalize and scale. The work connects approximation theory, routing, active parameters, and data geometry to explain why MoE models can improve efficiency while keeping strong predictive performance.
Research Explained
This section presents short and accessible explanations of selected research papers. Each note focuses on the motivation, main idea, method, results, and broader impact of the work.
Notes & Mini-Courses
I maintain a small collection of research notes, mini-courses, and technical explanations related to my research interests. These notes are intended for students, collaborators, and researchers interested in efficient machine learning, anomaly detection, optimization, and geometric approaches to AI.
Manifolds in Machine Learning
A short introduction to manifolds, the manifold hypothesis, representation learning, dimensionality reduction, and links with Machine learning.
Green AI and Efficient Deep Learning
Notes on computational cost, energy consumption, carbon footprint, compression, and efficiency-aware evaluation of learning systems.
Anomaly Detection in Time Series
A concise overview of anomaly detection settings, reconstruction-based methods, forecasting-based methods, and evaluation issues.
Selected publications
Publications are loaded automatically from HAL. See also my HAL profile and Google Scholar.
- Loading publications from HAL...
Teaching
- Introduction to Computer Science & Artificial Intelligence (Université Lumière Lyon 2)
- Introduction to Computer Science (SKEMA Business School)
- Management Information Systems (SKEMA Business School)
- Research project supervision (MSc Data Science & Artificial Intelligence, Université Côte d’Azur)