François Bachoc



Statistics, Uncertainty quantification, Kriging, Gaussian process, Covariance function estimation, Cross Validation, Maximum Likelihood, Computer model approximation and validation.

Current position

I am a Ph.D student in Statistics since October 2010 at the Commissariat à l'Energie Atomique et aux Energies Alternatives and at the Laboratoire de Probabilités et Modèles Aléatoires, Université Paris VII.

My Ph.D director is Josselin Garnier, Professor at Université Paris VII, and I am also supervised by Jean-Marc Martinez, Research Engineer at the Commissariat à l'Energie Atomique.

Research interests

My research interests are related to Gaussian process modeling. I work on the issue of the covariance function estimation. In this context, I study the Maximum Likelihood and Cross Validation estimators. I also study the influence of the irregularity of the sampling on the estimation, in an increasing-domain asymptotic context. On the subject of the covariance function estimation, I work in relation with the MASCOT NUM research group and the ReDICE project.

I am also particularly interested in the application of Gaussian process models to computer model validation, in a context where field data are available, and to surrogate-modeling of complex computer models. On the subject of computer model validation I participated in the Sub-Project 4: Model Validation & Calibration, Sensitivity and Uncertainty Quantification of the European NURISP project.