Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
This lecture by the instructor from King Abdullah University of Science and Technology focuses on the inference of reaction kinetics in the context of uncertainty quantification in combustion. The presentation covers topics such as rules inference, sensitivity analysis, parameter inference, and Bayesian inference. The lecture delves into the challenges posed by high-dimensional random parameter spaces and non-smooth behavior of quantities of interest. It also discusses the use of spectral representation and sparse quadrature methods for efficient analysis. The content includes detailed discussions on rate rule uncertainties, total sensitivities, sub-rule analysis, and the impact of individual subrules on the overall variability. The lecture concludes with insights on the hierarchical Bayesian framework for inferring Arrhenius law and the effect of measurement uncertainty on the inference process.