Quantifying Uncertainty for Machine Learning Based Diagnostic
This work addresses the need for reliable uncertainty estimates in safety-critical systems like particle accelerators, where predictions must be trustworthy for operational decisions.
The paper tackled the problem of uncertainty quantification for deep learning-based virtual diagnostics in particle accelerators, using ensemble methods and quantile regression neural networks to predict electron beam profiles with improved reliability.
Virtual Diagnostic (VD) is a deep learning tool 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 damaging the output. Given a prediction, it is necessary to relay how reliable that prediction is. This is known as 'uncertainty quantification' of a 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. 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.