Theme Team Metamodeling, Continuous Optimization and Applications (MOCA)


The MOCA team studies the management of numerical simulation models and their use in continuous optimization. Metamodeling encompasses here two types of "models of models" :

    • Statistical metamodels (e.g., Gaussian processes) that are built from input-output data. The data is often generated by the simulation of physical systems (e.g., with finite elements models). We specialize in statistical metamodels that are compatible with sparse data. Such metamodels typically underly optimization algorithms.

    • Metamodels that aim at producing computer programs by describing and processing a class of models. Such metamodels may also be metacodes that algorithmically describe a set of possible program instances.

The team has contributions in domains such as continuous optimization theory and computer experiments. In addition, the group implements simulation softwares for high-performance computing, discrete events simulation and numerical reproductibility.



  • Statistical models
  • Continuous optimization
  • Discrete event simulation
  • Scientific Computing
  • High Performance Computing
  • Numerical reproductibility

Last publications

Noé Lebreton, Julien Jacques, Julien Ah-Pine, Matthieu Neveu - Sept. 15, 2024
Pattern matching for multivariate time series forecasting
ENBIS-24 Conference

Amal Omrani, Anis Fradi, Chafik Samir - June 1, 2024
Reduced run-time and memory complexity regression with a Gaussian process prior

Alexis Chartrain, Gilles Dessagne, Noël Haddad, David R.C. Hill - May 29, 2024
Retrospective on the Digital Twin concept and perspectives for railways: the case of SNCF Réseau
2ème Congrès Annuel de la SAGIP (SAGIP '24)

Tanguy Appriou - May 14, 2024
Talk on Bayesian Optimization for High-Dimensional Problems
Journées CIROQUO mai 2024

Marc Grossouvre, Didier Rullière, Jonathan Villot - May 1, 2024
Predicting missing Energy Performance Certificates: Spatial interpolation of mixture distributions
Energy and IA

Tanguy Appriou - April 3, 2024
Talk on High-Dimensional Bayesian Optimization with a Combination of Kriging Models
Conférence MASCOT-NUM 2024

Hassan Maatouk, Didier Rullière, Xavier Bay - March 26, 2024
Large-scale constrained Gaussian processes for shape-restricted function estimation

Marc Grossouvre, Didier Rullière, Jonathan Villot - March 20, 2024
Enhancing buildings' energy efficiency prediction through advanced data fusion and fuzzy classification
Energy and Buildings

Benjamin A. Antunes, David R.C. Hill - March 19, 2024
Reproducibility, Replicability, and Repeatability: A survey of reproducible research with a focus on high performance computing

Hassan Maatouk, Didier Rullière, Xavier Bay - March 8, 2024
Efficient constrained Gaussian process approximation using elliptical slice sampling

All publications are here