IEMS 490: Special Topics: Uncertainty Quantification

Quarter Offered

Winter : TTH 3:30-4:50 ; Plumlee


 A course in Uncertainty Quantification with an emphasis on formulating and computation to extract predictions and uncertainty in computational and simulation models (including when real data supplements the model).  Main objectives include:

  • Organize uncertainty in the form of model parameters, inputs, discrepancy and stochastic error.
  • Design tools for uncertainty propagation, including samplers and other integration techniques.
  • Leverage statistical inference to infer on parameters and discrepancy using data.
  • Develop non-parametric learning for surrogate modeling to decrease reliance on computational requirements.
  • Construct predictive distributions for key quantities of interest.
  • Build verification and prediction accuracy assessments using scoring rules.


Optional texts include:

Uncertainty Quantification: Theory, Implementation and Applications (Smith)

Numerical methods for stochastic computations: a spectral method approach (Xui)

The Design and Analysis of Computer Experiments (Santner, Williams,  Notz)

Surrogates: Gaussian process modeling, design and optimization for the applied sciences (Gramacy)