Neural network biased corrections: Cautionary study in background corrections for quenched jets

TL;DR

This paper examines how neural network-based background corrections in jet measurements from heavy ion collisions are biased by jet quenching effects, potentially affecting the accuracy of jet property measurements.

Contribution

It provides a systematic study of biases introduced by neural network corrections due to jet quenching in heavy ion collisions using realistic simulations.

Findings

Neural network corrections are biased by jet quenching effects.

Biases impact the measurement of jet suppression factors.

Simulations show significant deviations in corrected jet properties.

Abstract

Jets clustered from heavy ion collision measurements combine a dense background of particles with those actually resulting from a hard partonic scattering. The background contribution to jet transverse momentum ($p_{T}$) may be corrected by subtracting the collision average background; however, the background inhomogeneity limits the resolution of this correction. Many recent studies have embedded jets into heavy ion backgrounds and demonstrated a markedly improved background correction is achievable by using neural networks (NNs) trained with aspects of jet substructure which are used to map measured jet $p_\mathrm{T}$ to the embedded truth jet $p_\mathrm{T}$. However, jet quenching in heavy ion collisions modifies jet substructure, and correspondingly biases the NNs' background corrections. This study investigates those biases by using simulations of jet quenching in central Au+Au collisions at $\sqrt{s_\mathrm{NN}}=200\;\mathrm{GeV}/c$ with hydrodynamically modeled quark-gluon plasma (QGP) evolution. To demonstrate the magnitude of the effect of such biases in measurement, a leading jet nuclear modification factor ($R_\mathrm{AA}$) is calculated and reported using the NN background correction on jets quenched utilizing a brick of QGP.