Papers

Amortized Bayesian Inference of GISAXS Data with Normalizing Flows

M Zhdanov, L Randolph, T Kluge… - arXiv preprint arXiv …, 2022 - arxiv.org
Computer Science paper cs.LG Suggest

… As the likelihood of the data-generating process is intractable, we employ simulation-based inference [11] via NFs. Due to the scarcity of experimental data, we train the model on …

Link to paper

BibTeX

@article{2210.01543v1,
Author        = {Maksim Zhdanov and Lisa Randolph and Thomas Kluge and Motoaki Nakatsutsumi and Christian Gutt and Marina Ganeva and Nico Hoffmann},
Title         = {Amortized Bayesian Inference of GISAXS Data with Normalizing Flows},
Eprint        = {2210.01543v1},
ArchivePrefix = {arXiv},
PrimaryClass  = {cs.LG},
Abstract      = {Grazing-Incidence Small-Angle X-ray Scattering (GISAXS) is a modern imaging
technique used in material research to study nanoscale materials.
Reconstruction of the parameters of an imaged object imposes an ill-posed
inverse problem that is further complicated when only an in-plane GISAXS signal
is available. Traditionally used inference algorithms such as Approximate
Bayesian Computation (ABC) rely on computationally expensive scattering
simulation software, rendering analysis highly time-consuming. We propose a
simulation-based framework that combines variational auto-encoders and
normalizing flows to estimate the posterior distribution of object parameters
given its GISAXS data. We apply the inference pipeline to experimental data and
demonstrate that our method reduces the inference cost by orders of magnitude
while producing consistent results with ABC.},
Year          = {2022},
Month         = {Oct},
Url           = {http://arxiv.org/abs/2210.01543v1},
File          = {2210.01543v1.pdf}
}

Share