Synchrotron-based X-ray scattering and spectroscopies are the state-of-the-art materials characterization method that measures materials’ structural and dynamical properties with an unprecedented level of precision. It plays a critical roles in a large variety of materials from catalysis to bio- macromolecules, from energy materials to quantum materials. In recent years, X-ray scattering has made significant progress thanks to the higher photon flux and multimodal measurements, which increases the data volume and data dimension. To address the associated challenge on data analysis, the X-ray scattering has also been augmented by computational modeling and the rapid implementation of machine learning. In light of this, we are proposing this timely Workshop to introduce the recent progresses how machine learning can benefit various perspectives of X-ray scattering and spectroscopies. We divide the Workshop into several sub-fields, including structural characterization from atomic to mesoscopic scales, dynamical and spectroscopic properties, and beam optimization and uni- and multimodal data analysis. In both structural and dynamical properties, we will balance the speakers from soft materials in chemistry and biology with speakers working on hard condensed matters. By focusing on a few emerging architectures of machine learning as well as highlighting the pressing questions that can be answered by machine learning, our proposed Workshop can tremendously benefit NSLS-II and CFN by bringing in potential new deployment on hardware and analysis augmentation, as well as new research opportunities. The workshop will bring together scientists in the field of X-ray scattering, materials informatics, and data science to seek for the viable pathways to augment NSLS-II beamlines to explore materials properties.
Workshop Organizers:
Mingda Li (MIT)
Deyu Lu (BNL)
Eli Stavitski (BNL)
Xiaohui Qu (BNL)