Artificial intelligence for online characterization of ultrashort X-ray free-electron laser pulses

Abstract

X-ray free-electron lasers (XFELs) as the world’s brightest light sources provide ultrashort X-ray pulses with a duration typically in the order of femtoseconds. Recently, they have approached and entered the attosecond regime, which holds new promises for single-molecule imaging and studying nonlinear and ultrafast phenomena such as localized electron dynamics. The technological evolution of XFELs toward well-controllable light sources for precise metrology of ultrafast processes has been, however, hampered by the diagnostic capabilities for characterizing X-ray pulses at the attosecond frontier. In this regard, the spectroscopic technique of photoelectron angular streaking has successfully proven how to non-destructively retrieve the exact time–energy structure of XFEL pulses on a single-shot basis. By using artificial intelligence techniques, in particular convolutional neural networks, we here show how this technique can be leveraged from its proof-of-principle stage toward routine diagnostics even at high-repetition-rate XFELs, thus enhancing and refining their scientific accessibility in all related disciplines.

Autonomous x-ray scattering

“…The data analysis step can also be automated in a variety of ways. The analysis pipeline can leverage existing analysis methods and software [51,62], or use a ML method optimized for the particular kind of x-ray data: scattering [63][64][65][66][67][68][69][70][71][72], diffraction [73][74][75][76][77][78][79] and crystallography [80,81], reflectivity [82][83][84], imaging [85][86][87][88][89][90][91], spectroscopy [92][93][94], photon correlation spectroscopy [95,96], or XFEL pulsed experiments [10,97,98] and single-particle diffraction [73,[99][100][101]. In all cases, analysis code must be wrapped in a workflow manager that can automatically trigger the analysis pipeline as soon as new data is collected.…”