Machine-learning-based jet momentum reconstruction in heavy-ion collisions

The precise reconstruction of jet transverse momenta in heavy-ion collisions is a challenging task. A major obstacle is the large number of (mainly) low-$p_{\mathrm{T}}$ particles overlaying the jets. Strong region-to-region fluctuations of this background complicate the jet measurement and lead to significant uncertainties. In this paper, a novel approach to correct jet momenta (or energies) for the underlying background in heavy-ion collisions is introduced. The proposed method makes use of common machine learning techniques to estimate the jet transverse momentum based on several parameters, including properties of the jet constituents. Using a toy model and HIJING simulations, the performance of the new method is shown to be superior to the established standard area-based background estimator. The application of the new method to data promises the measurement of jets down to extremely low transverse momenta, unprecedented thus far in data on heavy-ion collisions.

Figure 1

Illustration of the full proposed analysis strategy. Explanations are given in the text.

Figure 2

Residual distributions for several background estimators and $40\le p_{\mathrm{T},\mathrm{ch}\mathrm{jet}}^{\mathrm{true}}<60$ GeV/$c$.

Figure 3

Residual distributions for the neural network estimator and several (true) jet transverse momenta.

Figure 4

Comparison of residual distributions for several thermal background settings for $40\le p_{\mathrm{T},\mathrm{ch}\mathrm{jet}}^{\mathrm{true}}<60$ GeV/$c$.

Figure 5

Comparison of the widths of the residual distributions as a function of the jet resolution parameter for $40\le p_{\mathrm{T},\mathrm{ch}\mathrm{jet}}^{\mathrm{true}}<60$ GeV/$c$.

Figure 6

Comparison of residual distributions for gluon, quark, and inclusive jets for the neural network estimator.

Figure 7

Comparison of residual distributions for gluon, quark, and inclusive jets for the area-based method.

Figure 8

Comparison of residual distributions for jets created by JEWEL for the neural network estimator.

Figure 9

Reconstructed charged jet spectra in HIJING events and the ratio to ($N_{\mathrm{coll}}$-scaled) pythia jet spectra.

Figure 10

Jet mass distribution compared for different jet energy scales. A background correction for the jet mass itself was also applied [24].