Advanced Data Analytics in the Service of Structure Science, Convolutional Neural Networks for Space-Group Classification and beyond

events hall

Prof. Simon J L Billinge




Nanostructure is now at the center of many efforts to design advanced next-generation materials. Total scattering and atomic pair distribution function methods from x-ray synchrotron sources and the latest generation electron microscopes are especially valuable in this regard. The ability to measure PDFs rapidly with small beams of sufficiently high energy, together with modern detectors and integrated on the fly data analysis opens up completely new scientific possibilities but introduces huge data analysis and computational challenges. In this talk I will describe recent developments in the experimental methods to to study nanostructure, but with an emphasis on some of the emerging analysis methods that rely on artificial intelligence and machine learning to help scientists squeeze the most from their data.

Link to ZOOM