Freiburg 2019 – scientific programme

Sessions | Days | Selection | Search | Updates | Downloads | Help

FM: Fall Meeting

FM 82: Quantum & Information Science: Neural Networks, Machine Learning, and Artificial Intelligence III

FM 82.6: Talk

Thursday, September 26, 2019, 15:30–15:45, 3044

Reconstructing Nanoclusters from Single Wide-Angle Scattering Images with Neural Networks — •Thomas Stielow, Robin Schmidt, Thomas Fennel, and Stefan Scheel — Institut fürPhysik, Universität Rostock, Albert-Einstein-Straße 23, 18059 Rostock

Single-shot diffraction imaging by soft X-ray laser pulses is a valuable tool for structural analyses of unsupported and short-lived nanosystems, although inversion of the scattering patterns still prove challenging [1]. Deep learning, on the other hand, is widely used in data sciences for the extraction of information from images and sees more and more application in various sciences. We demonstrate how neural networks can be utilized in the reconstruction of objects from single-shot wide angle scattering patterns in the case of silver nanoclusters [2]. Our network is trained solely on data obtained by existing physical theories and can be applied to real-world experimental data with little to no prior knowledge of the specific experimental setup. With high quality real-time evaluation results, deep learning may hold the key for a fully automated analysis of scattering data and real-time reconstruction of ultrafast nanoscale dynamics probed at the next generation of X-ray light sources with high repetition rate.

[1] I. Barke et al., Nat. Comm. 6, 6187 (2015).
[2] T. Stielow et al., arXiv:1906.06883 (2019).