This seminar will discuss the usage of unfolding as applied to experiments in High Energy Particle physics experiments. A focus will be given to how these methods are applied to analyses, and how the treatment of uncertainties influences considerations such as the choice of algorithm and regularisation. An example statistical comparison of different unfolding techniques is included to help guide the choice of unfolding method and regularisation strengths. The included unfolding methods for the comparison are; Iterative Bayes, Truncated Singular Value Decomposition, Iterative dynamically stabilized, TUnfold and Gaussian Processes. Additionally, possible improvements and extensions are discussed.
Lydia Brenner is an experimental Particle Physicist at Nikhef, Amsterdam. She works on the ATLAS and FASER experiments at CERN, and has a special interest in the statistics behind data analysis methods and interpretations of results. She is the lead author of a recent paper on unfolding: https://arxiv.org/pdf/1910.14654.pdf
O. Behnke, L, Brenner, L. Lyons, N. Wardle, S. Algeri