About Me
I am a PhD candidate in Applied Mathematics working on probabilistic modeling and surrogate learning for complex mechanical systems. My thesis is carried out at the Université Polytechnique Hauts-de-France (UPHF) and École des Mines de Saint-Étienne (EMSE), within the ANR JCJC GAME project.
My research focuses on the use of Gaussian Processes (GPs) for the modeling, simulation and identification of mechanical systems governed by uncertain or incomplete data. A specific aspect of my work is the development of GP models with functional inputs, where input data are not simple vectors but continuous fields such as wear topographies or flow distributions.
The aim is to build efficient surrogate models to replace computationally expensive solvers (e.g. finite element codes), while ensuring predictive accuracy and uncertainty quantification. Applications include fluid flow in porous media and contact-induced phenomena such as brake pad wear and acoustic pressure modeling.
Beyond forward prediction, my research also tackles inverse problems, where the goal is to reconstruct input fields from observable mechanical responses—enabling, for instance, the detection of critical scenarios or degradation conditions.
I am committed to producing open-source tools in Python and R, contributing to reproducibility and interdisciplinary collaboration. Through this work, I aim to advance interpretable and uncertainty-aware modeling in engineering and computational mechanics.
Contact
Université Polytechnique Hauts-de-France
Département de Mathématiques (DMATHS), Bât. Abel de Pujol 2
59313 Valenciennes Cedex 9, France
Email: RazakChristophe.SabiGninkou@uphf.fr
Other: razsabigninchrist@gmail.com / sabigninkou.razakchristophe@imsp-uac.org
PhD Thesis Information
Title: Gaussian Process Modeling of Mechanical Random Fields
Supervisors: R. Le Riche, Y. Maïnassara, A. Lopez-Lopera, F. Massa, L. Reding
Funding: ANR JCJC GAME
Research Interests
- Gaussian Processes
- IHigh-computational-cost code
- Machine Learning
- Multi-task learning
- Uncertainty Quantification
- Functional Data Analysis
- Inverse Problems
Certifications
Below is a list of certifications I have completed, covering topics in deep learning, AI, and scientific writing.
-
Improving Deep Neural Networks
DeepLearning.AI | Verify | PDF -
Structuring Machine Learning Projects
DeepLearning.AI | Verify | PDF -
Neural Networks and Deep Learning
DeepLearning.AI | Verify | PDF -
Responsible AI: Applying AI Principles
Google Cloud | Verify | PDF -
Introduction to Generative AI
Google Cloud | Verify | PDF -
Introduction to Responsible AI
Google Cloud | Verify | PDF -
Introduction to Large Language Models
Google Cloud | Verify | PDF -
How to Write and Publish a Scientific Paper
École Polytechnique | Verify | PDF