Assistant professor, Lumière University Lyon 2, Claude Bernard University Lyon 1, ERIC, France
Dr. Anis Fradi is a dedicated academic and researcher in the fields of computer science and applied mathematics, currently serving as an Associate Professor at Université Lumière Lyon 2, France. With a strong interdisciplinary background and a passion for machine learning, optimization, and Bayesian inference, he brings a wealth of experience in developing efficient, interpretable models for high-dimensional and structured data. His work bridges theoretical foundations and practical applications, especially in areas like image classification, regression models, and manifold-valued data analysis. 🇫🇷💻📊
Publication Profile
🎓 Education Background
Dr. Fradi earned a dual PhD in Computer Science from Université Clermont Auvergne, France, and Applied Mathematics from the University of Monastir, Tunisia (2017–2021). His thesis focused on Bayesian Inference in 2D and 3D Shape Analysis. He also holds a Research Master’s Degree in Mathematics and Applications (2013–2015, University of Sousse) with honors and a Bachelor’s Degree in Mathematics and Applications (2010–2012) from the same university. His academic journey reflects a solid foundation in mathematical modeling and algorithmic development. 📘📐👨🎓
💼 Professional Experience
Dr. Fradi began his career as a lecturer in Tunisia before transitioning to multiple academic roles in France. He has served as a Postdoctoral Researcher at CNRS-LIMOS and Inria Bordeaux – Sud-Ouest, focusing on learning on manifolds and probabilistic representations. Between 2023 and 2024, he was a Temporary Lecturer and Research Assistant at Université Clermont Auvergne. Since September 2024, he has held a permanent Associate Professorship at Université Lumière Lyon 2, contributing to both teaching and research in computer science and data mining. 🏫📈🧠
🏅 Awards and Honors
Dr. Fradi has been recognized for his impactful contributions to AI and data science. He received the Best Paper Award at PRICAI 2023 in Jakarta and the prestigious CNRS 80|Prime Award, a competitive French national research incentive. These accolades highlight his innovative work in robust AI models and inference techniques. 🏆🎖️📚
🔍 Research Focus
His research revolves around Bayesian learning, manifold-valued data, Gaussian processes, optimization, and interpretable AI. He aims to reduce computational complexity while maintaining model accuracy and robustness. His work is especially prominent in image classification, regression models with low complexity, and analysis on non-Euclidean spaces such as Riemannian manifolds. 🧮🤖🌐
🔚 Conclusion
Dr. Anis Fradi stands out as a thought leader blending advanced mathematical concepts with modern AI to solve complex real-world problems. His career is marked by interdisciplinary excellence, international collaborations, and a commitment to both innovation and education in data science and machine learning. 🚀📚🌍
📚 Top Publications
ConvKAN: Towards Robust, High-Performance and Interpretable Image Classification – 2025, Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
Cited by: Early impact; award-nominated contribution in interpretable AI and convolutional kernel adaptive networks.
Decomposed Gaussian Processes for Efficient Regression Models with Low Complexity – 2025, Entropy (MDPI)
Cited by: Researchers in scalable Bayesian models and metamodeling; rapidly gaining attention.
A New Bayesian Approach to Global Optimization on Parametrized Surfaces in R3\mathbb{R}^3R3 – 2024, Journal of Optimization Theory and Applications
Cited by: Mathematicians and data scientists working on optimization and 3D surface modeling.
A New Framework for Evaluating the Validity and the Performance of Binary Decisions on Manifold-Valued Data – 2024, Book Chapter
Cited by: Scholars focusing on non-Euclidean data analysis and manifold learning.
Reduced Run-Time and Memory Complexity Regression with a Gaussian Process Prior – 2024, Conference Paper (HAL)
Cited by: Applications in real-time systems and efficient predictive modeling.