Dates: July 23-July 27, 2018.
The emerging area of uncertainty quantification (UQ) for computer models is the focus of the 2018-2019: Program Model Uncertainty: Mathematical and Statistical (MUMS) at the Statistical and Applied Mathematical Sciences Institute (SAMSI). The aim of the SAMSI program is to bring together statistical and mathematical scientists to tackle important common research problems in UQ and to train the next generation of UQ researchers. To this end, the Canadian Statistical Sciences Institute (CANSSI) and SAMSI are jointly sponsoring the Summer School on UQ.
The objective of this 5-day summer school is to present the statistical and mathematical foundations for UQ in a wide variety of applications. Among the many challenges include understanding all of the sources of variability (statistical and mathematical) and combining models and data to make inferences about physical systems, or the parameters that govern them. The approach will be one where attendees will learn UQ methods by implementation and where roughly half of the time will be spent on problem solving (exercises and the real-world problem). We also aim to establish a common nomenclature among the mathematical and statistical sciences communities.
The target audience for the summer school is graduate students and post-docs interested in learning about UQ. Senior level undergraduate students are also welcome, particularly those who have interest, or experience, in UQ and also implementing new methods (i.e., you should know some computing).
Note: Summer School participants do NOT have to be involved in the SAMSI program.
|Derek Bingham||Simon Fraser University|
|Paul Constantine||University of Colorado, Boulder|
|Leanna House||Virginia Tech|
|Dave Higdon||Social Decisions Analytics Lab, Virginia Tech|
We aim to have the first half of each day consist of presentations and lectures and the second half of the day consist of work on activities, worksheets, numerical experiments and coding. We will take great care to properly distinguish between applied problems, concepts and methods. It is essential to first formulate the problem(s) and then to discuss/compare different potential approaches – with emphasis on those that will be explored in the SAMSI program. Participants will gain hands-on experience using new and existing code for cutting edge techniques (e.g., Bayesian computer model calibration and polynomial chaos). Likely coding environments are Matlab, Python and R. The summer school participants will be introduced to a real research problem.
Our aim is to provide accommodation for out of town graduate students and post-docs. We will be making reservations for you at one of two hotels: (i) Best Western Plus Coquitlam; or (ii) Executive Plaza Coquitlam Please indicate that a reservation is required when you register. Do not contact the hotels directly. Spaces are limited, and will we do our best… no guarantees. Please register before May 31, 2018 if you plan to request accommodations.