MA 797 Uncertainty Quantification

Semester: Spring 2014

Days and Times: Tuesdays and Thursdays, 11:45-1:00

Instructor: Ralph Smith

Topics: The field of uncertainty quantification is evolving rapidly due to increasing emphasis on models that have quantified uncertainties for large-scale applications, novel algorithm development, and new computational architectures that facilitate implementation of these algorithms. In this course, we develop the basic concepts, theory, and algorithms necessary to quantify input and response uncertainties for a variety of simulation models. Examples will be drawn from applications including weather and climate models, subsurface hydrology and geology models, nuclear power plant design, and models for biological phenomena. The topics will include concepts from probability and statistics, parameter selection techniques, frequentist and Bayesian model calibration, propagation of uncertainties, quantification of model discrepancy, surrogate model construction, and local and global sensitivity analysis.

This interdisciplinary course is designed for students in mathematics, statistics, engineering, operations research, and the sciences. A basic knowledge of probability, linear algebra, ordinary and partial differential equations, and introductory numerical analysis techniques is assumed. Coursework will consist of projects.

Contact Ralph Smith (rsmith _at_ for more details.