A simultaneous unbinned differential cross section measurement of twenty-four $Z$+jets kinematic observables with the ATLAS detector

TL;DR

This paper presents a novel unbinned, simultaneous measurement of twenty-four Z+jets observables at the LHC using machine learning, enabling flexible analysis and improved understanding of QCD phenomena.

Contribution

It introduces a new unbinned, particle-level measurement of multiple Z+jets observables using OmniFold, surpassing previous binned approaches and allowing for versatile future analyses.

Findings

First unbinned measurement of 24 Z+jets observables

Enhanced data reusability and analysis flexibility

Provides detailed dataset for QCD studies

Abstract

$Z$ boson events at the Large Hadron Collider can be selected with high purity and are sensitive to a diverse range of QCD phenomena. As a result, these events are often used to probe the nature of the strong force, improve Monte Carlo event generators, and search for deviations from Standard Model predictions. All previous measurements of $Z$ boson production characterize the event properties using a small number of observables and present the results as differential cross sections in predetermined bins. In this analysis, a machine learning method called OmniFold is used to produce a simultaneous measurement of twenty-four $Z$+jets observables using $139$ fb$^{-1}$ of proton-proton collisions at $\sqrt{s}=13$ TeV collected with the ATLAS detector. Unlike any previous fiducial differential cross-section measurement, this result is presented unbinned as a dataset of particle-level events, allowing for flexible re-use in a variety of contexts and for new observables to be constructed from the twenty-four measured observables.