Distinguishing W′ signals at hadron colliders using neural networks

Abstract

We investigate a neural-network-based hypothesis test to distinguish different W′ and charged scalar resonances through the ℓ+ET channel at hadron colliders. This is traditionally challenging due to a fourfold ambiguity at proton-proton colliders, such as the Large Hadron Collider. Of the neural network approaches we study, we find a multiclass classifier based on a fully connected neural network trained upon two-dimensional histograms made from kinematic variables of the final state ℓ to be the most powerful. Furthermore, by considering the one-jet processes, we demonstrate that one can generalize to multiple two-dimensional histograms to represent different variable pairs. Finally, as a comparison to traditional approaches, we compare our method with Bayesian hypothesis testing and discuss the pros and cons of each approach. The neural network scheme presented in this paper is a powerful tool that can help probe the properties of charged resonances.