Machine learning enabling high-throughput and remote operations at large-scale user facilities
Bruce D. Ravel, Tatiana Konstantinova, Phillip Michael Maffettone, Stuart Campbell, Andi Barbour, Daniel Olds
Imaging, scattering, and spectroscopy are fundamental in understanding and discovering new functional materials. Contemporary innovations in automation and experimental techniques have led to these measurements being performed much faster and with higher resolution, thus producing vast amounts of data for analysis. These innovations are particularly pronounced at user facilities and synchrotron light sources. Machine learning (ML) methods are regularly developed to interpret large datasets in real-time with measurements. However, there remain conceptual barriers to entry for the domain experts, and technical barriers for deploying ML models. Herein, we demonstrate a variety of archetypal ML models for on-the-fly analysis at multiple beamlines at the National Synchrotron Light Source II (NSLS-II). We describe these examples instructively, with a focus on integrating the models into experimental workflows, such that the reader (or facility user) can easily integrate their own ML techniques into experiments at NSLS-II or facilities with common infrastructure. The framework presented here shows how diverse ML models for offline use can operate in conjunction with feedback loops via integration into the existing Bluesky Suite for experimental orchestration and data management with little effort.
, Konstantinova, T.
, Maffettone, P.
, Campbell, S.
, Barbour, A.
and Olds, D.
Machine learning enabling high-throughput and remote operations at large-scale user facilities, Digital Discovery, [online], https://doi.org/10.1039/D2DD00014H, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=934081
(Accessed September 30, 2023)