Uncertainty quantification for ptychography using normalizing flows
This work addresses uncertainty quantification for ptychography, which is important for researchers in material science and imaging, but it appears incremental as it applies an existing deep learning method to a new domain.
The paper tackled the problem of uncertainty quantification in ptychography, a material characterization technique, by using normalizing flows to model the high-dimensional posterior, enabling uncertainty characterization for reconstruction quality assessment and artifact detection. The method was demonstrated on synthetic and experimental data, though no concrete performance numbers were provided.
Ptychography, as an essential tool for high-resolution and nondestructive material characterization, presents a challenging large-scale nonlinear and non-convex inverse problem; however, its intrinsic photon statistics create clear opportunities for statistical-based deep learning approaches to tackle these challenges, which has been underexplored. In this work, we explore normalizing flows to obtain a surrogate for the high-dimensional posterior, which also enables the characterization of the uncertainty associated with the reconstruction: an extremely desirable capability when judging the reconstruction quality in the absence of ground truth, spotting spurious artifacts and guiding future experiments using the returned uncertainty patterns. We demonstrate the performance of the proposed method on a synthetic sample with added noise and in various physical experimental settings.