Reconstruction of Angular Correlations in Associated Top Quark and Dark Matter Mediator Production

For the process of single top quark production within the ‘‘simplified model’’ with a scalar dark matter mediator, a new variable based on angular correlations was presented, for the proper reconstruction of which it is necessary to separate the contributions of two undetectable particles: the neutrino and the mediator. In this work, various machine learning approaches for reconstructing the momenta of these particles are analyzed. A comparison is made between the results obtained using a multilayer perceptron and the Normalizing Flows architectures. The neural networks based on Normalizing Flows, presented in this work, demonstrate a high quality of reconstruction of the target variable and can be used for collider data analysis when applying the unfolding procedure to restore parton-level correlations.