Amortized Bayesian Inference of GISAXS Data with Normalizing Flows
This work addresses a domain-specific problem for material researchers by providing a faster inference method for GISAXS data analysis, though it appears incremental as it builds on existing simulation-based and normalizing flow techniques.
The paper tackles the ill-posed inverse problem of reconstructing nanoscale material parameters from GISAXS data, which is traditionally slow with methods like ABC, and demonstrates that their method using variational auto-encoders and normalizing flows reduces inference cost by orders of magnitude while maintaining consistency with ABC.
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.