ECOS Nord projet C25M01 (France-Colombia) – Large-scale probabilistic machine learning models for aerodynamic data fusion

This 3-year project (starting in January 2025) is funded by the 2024 ECOS Nord call, as part of the joint research program France-Colombia.

Research team

• French team
  • A. F. López-Lopera (Co-PI), CERAMATHS, UPHF
  • N. Bartoli, ONERA
  • F. Massa, LAMIH, UPHF
  • M. Castaño-Aguirre, ONERA, CERAMATHS-LAMIH, UPHF
• Colombian team
  • H. F. Garcia-Arias (Co-PI), UdA
  • A. E. Salazar-Jimenez, UdA
  • J. F. Vargas-Bonilla, UdA
  • C. Guarnizo, ITM

Summary

This project seeks to establish machine learning models that can account for large multi-source datasets and uncertainty quantification. The main interest are:

• Large-scale probabilistic models for multi-fidelity databases.
• Inference methods for parameter estimation of the multi-fidelity models.
• Applications in aerodynamic data fusion.

Keywords. Aerodynamic codes, probabilistic modeling and simulation, uncertainty quantification, multi-fidelity, machine learning, Gaussian processes.

To acknowledge this project: “This work was conducted as part of the ECOS Nord project C25M01.”