ARU - towards automatic unfolding of detector effects

This article presents the ARU algorithm, a general non-interactive algorithm for the unfolding of detector effects (resolution effects, efficiency, non-linear response) from one-dimensional data distributions. ARU uses an unbinned maximum-likelihood fit with a weighted regularization term, based on the relative information in the solution with respect to a reference distribution. The optimal regularization weight is found by minimizing the mean squared error of the solution. The algorithm’s performance is demonstrated in a study of a toy data sets. The analysis shows that the bias on average is smaller than the statistical uncertainties which are properly estimated by the fit.