Uncertainty Quantification for Virtual Diagnostic of Particle Accelerators

TL;DR

This paper explores methods to quantify uncertainty in deep learning-based virtual diagnostics for particle accelerators, enhancing the reliability of predictions in safety-critical applications.

Contribution

It introduces ensemble methods and quantile regression neural networks to assess prediction uncertainty in experimental accelerator data.

Findings

Ensemble methods improve uncertainty estimation accuracy.

Quantile regression neural networks provide reliable confidence intervals.

Enhanced uncertainty quantification aids safe decision-making in accelerator diagnostics.

Abstract

Virtual Diagnostic (VD) is a computational tool based on deep learning that can be used to predict a diagnostic output. VDs are especially useful in systems where measuring the output is invasive, limited, costly or runs the risk of altering the output. Given a prediction, it is necessary to relay how reliable that prediction is, i.e. quantify the uncertainty of the prediction. In this paper, we use ensemble methods and quantile regression neural networks to explore different ways of creating and analyzing prediction's uncertainty on experimental data from the Linac Coherent Light Source at SLAC National Lab. We aim to accurately and confidently predict the current profile or longitudinal phase space images of the electron beam. The ability to make informed decisions under uncertainty is crucial for reliable deployment of deep learning tools on safety-critical systems as particle accelerators.