Comparison of Machine Learning Approach to other Unfolding Methods
Petr Baron
Unfolding in high energy physics represents the correction of measured spectra in data for the finite detector efficiency, acceptance, and resolution from the detector to particle level. Compared to other commonly used unfolding methods, recent machine learning approaches provide unfolding on an event-by-event basis using all the information from the collision similarly to the face recognition problem and allows to unfold spectra in a continuous way (independent of binning choice) the simultaneous unfolding of a large number of variables and thus can cover a wider region of the features that effect detector response. This study focuses on a simple comparison of commonly used methods in RooUnfold package to the machine learning package Omnifold.