A search for leptonic photon Z_l at all three CLIC energy stages by using artificial neural networks (ANN)

TL;DR

This paper explores the potential to detect a hypothetical leptonic photon Z_l at CLIC collider energies using neural networks to estimate cross-sections, demonstrating machine learning's applicability in particle physics predictions.

Contribution

It introduces a method combining neural networks with theoretical calculations to predict Z_l detection prospects at CLIC energies, extending previous analyses.

Findings

Z_l can be observed at CLIC energies if coupling g_l ≥ 10^{-3}

Neural networks effectively estimate cross-sections for Z_l production

Predictions support machine learning as a tool in collider physics

Abstract

In this work, the possible dynamics of the massive leptonic photon \(Z_{l}\) are reconsidered via the process \(e^{+}e^{-} \ rightarrow \mu^{+}\mu^{-}\) at Compact Linear Collider (CLIC) with updated center of mass energies (\(380\ GeV,\ 1500\ GeV\ and\ 3000\ GeV\)). We show that the new generation colliders as CLIC can observe massive leptophilic vector boson Z_l with mass up to the center of mass energy, provided that leptonic coupling constant is \(g_{l} \geq 10^{- 3}\). In this study, we also estimated the cross-sections by artificial neural networks using the theoretical results we obtained for CLIC. According to the results obtained, it was seen that these predictions could be made through machine learning.