Conference paper
Adaptive spectral tensor-strain decomposition for the construction of surrogate models
The construction of surrogate models is important as a mean of acceleration in computational methods for uncertainty quantification (UQ). When the forward model is particularly expensive, surrogate models can be used for the forward propagation of uncertainty [4] and the solution of inference problems [5].
An adaptive construction is necessary to meet the prescribed accuracy tolerances with the lowest computational effort.
Language: | English |
---|---|
Year: | 2015 |
Proceedings: | SIAM Conference on Computational Science and Engineering (SIAM CSE 2015) |
Types: | Conference paper |
ORCIDs: | Engsig-Karup, Allan Peter |