Applied Mathematics Colloquium

Tuesday, November 8, 2016
2:45 PM - 3:45 PM
Add to Calendar

Link added to clipboard:

https://events.columbia.edu/cal/event/eventView.do?b=de&calPath=%2Fpublic%2Fcals%2FMainCal&guid=CAL-00bbdcc7-56cbff19-0156-dc4dc7ba-00006701events@columbia.edu&recurrenceId=
Lindley Graham
Department of Scientific Computing at Florida State University

"Measure-Theoretic Methods for Uncertainty Quantification"
We discuss how uncertainty quantification problems can benefit from being solved within a measure-theoretic framework. Forward uncertainty quantification and inverse uncertainty quantification both often involve the estimation of parameters or statistical quantities of complex computations models via sample based integration. This estimation can be very expensive due the large number of samples required. When the model is sufficiently complex, sample based integration methods such as the Monte Carlo method may be the only feasible approach. In this talk we will also address some advantages and insights gained from reinterpreting sample based integration within a measure-theoretic framework.
Event Contact Information:
Christina Rohm
212 854 1586
[email protected]
LOCATION:
  • Morningside
TYPE:
  • Lecture
CATEGORY:
  • Engineering
EVENTS OPEN TO:
  • Alumni
  • Faculty
  • Public
  • Staff
  • Student
BACK TO EVENTS

Date Navigation Widget

Filter By

Subscribe Export Options

Getting to Columbia

Other Calendars

Guests With Disabilities