Conference paper
Spectral Tensor-Train Decomposition for low-rank surrogate models
The construction of surrogate models is very important as a mean of acceleration in computational methods for uncertainty quantification (UQ). When the forward model is particularly expensive compared to the accuracy loss due to the use of a surrogate – as for example in computational fluid dynamics (CFD) – the latter can be used for the forward propagation of uncertainty [7] and the solution of inference problems.
Language: | English |
---|---|
Year: | 2014 |
Proceedings: | Spatial Statistics and Uncertainty Quantification on Supercomputers |
Types: | Conference paper |
ORCIDs: | Engsig-Karup, Allan Peter |