SCCS Colloquium - Jul 8, 2020

From Sccswiki
Jump to navigation Jump to search
Date: July 8, 2020
Room: Online (password: SCCS)
Time: 15:00 - 16:00

Anastasiya Liatsetskaya: Adaptive Quadrature with the Combination Technique for UQ Applications

Bachelor's thesis submission talk. Anastasiya is advised by Michael Obersteiner.

The core of this Bachelor's Thesis is comparison of results computed by the Generalized spatially adaptive Sparse Grid Combination Technique with dimension-wise refinement, developed by Michael Obersteiner and Hans-Joachim Bungartz, and the results computed by non-adaptive quadrature rules. The function f(x), for which the weighted integral is approximated, represents the values of numerical simulation containing uncertain parameters. According to the results the Quasi Monte Carlo quadrature rule has shown the best results and the adaptive quadrature rule has shown similar values with Gaussian quadrature rule in estimated error decay rate and estimated error values.

Keywords: Adaptive Quadrature, Combination Technique, Uncertainty Quantification