Search for Single Top Quark Production at the D0 Experiment using Bayesian Neural Networks

TL;DR

This paper details a methodology employing Bayesian neural networks to distinguish single top quark production signals from background noise, leading to the first evidence of such production at D0 in Fermilab Tevatron data.

Contribution

It introduces the use of Bayesian neural networks for analyzing single top quark production, demonstrating their effectiveness in high-energy physics data analysis.

Findings

First evidence of single top quark production at D0

Bayesian neural networks effectively separate signal from background

Methodology improves detection sensitivity in collider experiments

Abstract

We present the methodology used to measure the single top quark production cross section in the D0 experiment, and show as an example the results that led to the first evidence of single top quark production in D0 at the Fermilab Tevatron proton-antiproton collider. The selected events are mostly backgrounds, which we separate from the expected signals using three multivariate analysis techniques, one of them being Bayesian neural networks, which we will describe here.