m (Scipediacontent moved page Draft content 557152526 to Mett et al 2025a) |
|||
| Line 1: | Line 1: | ||
| + | ==Abstract== | ||
| + | |||
| + | We investigate the use of normalizing flows to approximate transport maps from tractable reference densities to complex Bayesian posterior distributions for Bayesian model updating. A Gaussian process (GP) surrogate with active sampling is used to provide a differentiable target density for optimizing the transport map. While results show normalizing flows can capture multimodal behavior in a simple example, further work is needed to refine the active sampling strategy and enable mode identification in the GP surrogate for robust multimodal density approximation. | ||
| + | |||
== Full Paper == | == Full Paper == | ||
| − | <pdf>Media: | + | <pdf>Media:Mett_et_al_2025a_5518_I250452_23_QK9T.pdf</pdf> |
Latest revision as of 08:34, 27 June 2025
Abstract
We investigate the use of normalizing flows to approximate transport maps from tractable reference densities to complex Bayesian posterior distributions for Bayesian model updating. A Gaussian process (GP) surrogate with active sampling is used to provide a differentiable target density for optimizing the transport map. While results show normalizing flows can capture multimodal behavior in a simple example, further work is needed to refine the active sampling strategy and enable mode identification in the GP surrogate for robust multimodal density approximation.
Full Paper
Document information
Published on 16/05/25
Submitted on 16/05/25
Volume Recent advances in Bayesian computation for uncertainty-aware inverse analysis, 2025
DOI: 10.23967/icossar.2025.053
Licence: CC BY-NC-SA license
Share this document
claim authorship
Are you one of the authors of this document?