Gaussian processes for autonomous data acquisition at large-scale synchrotron and neutron facilities

The execution and analysis of complex experiments are challenged by the vast dimensionality underlying parameter spaces. Although an increase in data-acquisition rates should allow broader querying space, complexity subtle dependence model function on input parameters remains daunting owing to sheer number variables. New strategies for autonomous data acquisition being developed, with one promising direction use Gaussian process regression (GPR). GPR is a quick, non-parametric robust approximation uncertainty quantification method that can be applied directly acquisition. We review GPR-driven experimentation illustrate its functionality using real-world examples from large experimental facilities USA France. introduce basics loop focus processes, then shift infrastructure needs built around create closed loop. Finally, case studies we discuss show Gaussian-process-based widely applicable facilitate optimal instruments enabling efficient high-value datasets. (GPR) powerful, technique This Review introduces discusses several cases different fields.