My research is mainly focused on Bayesian sequential and batch-sequential evaluation strategies of functions using Gaussian process models. This includes: 

  • Kriging-based parrallel optimization and inversion
  • Stepwise Uncertainty Reduction (SUR) strategies
  • Gaussian process simulation

Keywords: Design of computer experiments, Gaussian process models, batch-sequential strategies, inversion, global optimization, active learning, Stepwise Uncertainty Reduction.

I also have fruitful collaborations with the climate impact group (GIUB) and the Oeschger center for climate change research  of the University of Bern.

In the past, I received support from the ReDice consortium, the French Institute of Nuclear Safety (IRSN), the University of Bern, and the University of Zurich

Publications (Updated in Oct. 2018):

Journal Articles in Computer Experiments / Machine Learning / Bayesian Optimization / Bayesian Inversion:

  • D. Rullière, N. Durrande, F. Bachoc, C. Chevalier. Nested Kriging predictions for datasets with a large number of observations. Statistics and Computing, vol. 28(4), pp 849-867 (2018). Article Preprint
  • D. Azzimonti, J. Bect, C. Chevalier, D. Ginsbourger. Quantifying uncertainties on excursion sets under a Gaussian random field priorSIAM/ASA J. Uncertainty Quantification, vol. 4(1), pp 850-874 (2016)Article Preprint
  • C. Chevalier, X. Emery, D. Ginsbourger. Fast Update of Conditional Simulation Ensembles. Mathematical Geosciences, vol. 47(7), pp 771-789 (2015). Article Preprint
  • D. Ginsbourger, J. Baccou, C. Chevalier, F. Perales, N. Garland, Y. Monerie. Bayesian adaptive reconstruction of profile optima and optimizers. SIAM/ASA J. Uncertainty Quantification, vol. 2, pp 490-510 (2014). Article Preprint
  • C. Chevalier, J. Bect, D. Ginsbourger, E. Vazquez, V. Picheny, Y. Richet. Fast parallel kriging-based stepwise uncertainty reduction with application to the identification of an excursion set. Technometrics, vol. 56(4), pp 455-465 (2014). Article Preprint
  • C. Chevalier, V. Picheny, D. Ginsbourger. Kriginv: An efficient and user-friendly implementation of batch sequential inversion strategies based on kriging. Computational Statistics & Data Analysis, vol. 71, pp 1021-1034 (2014). Article Preprint 


Conference Proceedings (peer reviewed):

  • D. Ginsbourger, J. Baccou, C. Chevalier, F. Perales. Design of computer experiments using competing distances between set-value inputs. mODa 11, Advances in Model-Oriented Design and Analysis, Contributions to Statistics, pp 123-131 (2016). Article
  • S. Marmin, C. Chevalier, D. Ginsbourger. Differentiating the multipoint Expected Improvement for optimal batch design. In P. Pardalos et al. (Eds.) Machine Learning, Optimization, and Big Data, Lecture Notes in Computer Science 9432, pp 37-48 (2016). Article Preprint
  • C. Chevalier, D. Ginsbourger, X. Emery. Corrected kriging update formulae for batch-sequential data assimilation. In Pardo-Igúzquiza, E., et al. (Eds.) Mathematics of Planet Earth, pp 119-122 (2014). Article Preprint 
  • C. Chevalier, D. Ginsbouger, J. Bect, I. Molchanov. Estimating and quantifying uncertainties on level sets using the vorob’ev expectation and deviation with gaussian process modelsmODa 10, Advances in Model-Oriented Design and Analysis, Contributions to Statistics, pp 35-43 (2013). Article Preprint
  • C. Chevalier, D. Ginsbouger. Fast computation of the multi-points expected improvement with applications in batch selection. In G. Nicosia and P. Pardalos (Eds.) Learning and Intelligent Optimization, Lecture Notes in Computer Science, pp 59-69, Springer (2013). Article Preprint


Journal Articles in Statistics in Climate Sciences:

  • O. Martius, S. Pfahl, C. Chevalier. A global quantification of compound precipitation and wind extremes. Geophysical research letters, vol. 43, pp 7709-7717 (2016). Article
  • Y. Barton, P. Giannakaki, H. von Waldow, C. Chevalier, S. Pfahl, O. Martius. Clustering of Regional-scale Extreme Precipitation Events in Southern Switzerland. Monthly Weather Review 144, pp 347-369 (2015). Article
  • P. Froidevaux, J. Schwanbeck, R. Weingartner, C. Chevalier, O. Martius. Flood triggering in Switzerland: the role of daily to monthly preceding precipitation. Hydrology and Earth System Sciences Discussions, vol. 19(9), pp 3903-3924 (2015). Article
  • I. Mahlstein, O. Martius, C. Chevalier, D. Ginsbourger. Changes in the odds of extreme events in the Atlantic basin depending on the position of the extratropical jet. Geophysical research letters, vol. 39, (2012). Article


Articles in review:

  • C. Chevalier, O. Martius, D. Ginsbourger. Modeling non-stationary extreme dependence with stationary max-stable processes and multidimensional scaling. Status: Submitted.
  • D. Azzimonti, D. Ginsbourger, C. Chevalier, J. Bect, Y. Richet. Adaptive design of experiments for conservative estimation of excursion sets. Status: Accepted with revision. Preprint
  • S. Marmin, C. Chevalier, D. Ginsbourger. Efficient batch-sequential Bayesian optimization with moments of truncated Gaussian vectors. Status: In revision. Preprint

Selection of invited Talks and conferences (last Update: Feb. 2017):

  • Apr 2016, SIAM conference on Uncertainty Quantification: Stepwise Uncertainty Reduction strategies for Inversion of Expensive Functions with Nuisance Parameters. Lausanne, Switzerland.

  • Nov 2015, Workshop on Big Data and Computer Experiments. Toulouse, France.

  • Nov 2014, GRF-Sim Workshop: Fast Update of Conditional Simulation Ensembles. Bern, Switzerland.

  • Nov 2014, Invited talk for the Institute of Geography at the University of Bern: Spatial modeling of extreme rainfall in Switzerland. Bern, Swizerland.

  • Jul. 2014, Invited talk, UCM 2014 Conference: Gaussian process update formulas for uncertainty quantification and computer experiments, Sheffield, UK.
  • Jun. 2014, Joint Research Conference: Bayesian adaptive construction of profile optima and optimizers, Seattle, USA.
  • Apr. 2014, Chorus Workshop: "Kriging and Gaussian processes for computer experiments". Presentation on Fast Update of Conditional Simulation Ensembles. Institut Henri Poincaré, Paris, France.
  • Apr. 2014, Presentation of the ReDICE consortium for the GDR Mascot Num. ETH Zurich, Switzerland.
  • Jan. 2014,  Invited talk, Fast sequential evaluation strategies using Gaussian process models, University of Zurich, Switzerland.
  • Nov. 2013, Invited talk, Parallel optimization using the Multi-points Expected Improvement. Dortmund, Germany.
  • Sept 2013, Ph.D. defense, Bern, Switzerland.
  • Jan. 2013, LION7 conference: Fast computation of the Multi-points Expected Improvement with application in Batch selection.  Catania, Italy.
  • Jul. 2012, UCM 2012 conference: Multipoint sampling criteria for the identification of excursion sets. Sheffield, UK.
  • Apr. 2012, SIAM conference on Uncertainty Quantification: Quantifying and Reducing Uncertainties on a set of failure using Random Set theory and Kriging. Railey, USA.
  • Dec. 2011, ERCIM'11 conference: KrigInv, an R package for sequential inversion of expensive-to-evaluate black-box simulators. London, UK