Overview: Jet quenching with machine learning

TL;DR

This review discusses how machine learning techniques are applied to analyze jet quenching phenomena in heavy ion collisions, aiming to improve jet reconstruction, classification, and understanding of quark-gluon plasma properties.

Contribution

It summarizes recent advances in applying machine learning to jet quenching studies, including jet reconstruction, classification, and origin identification in heavy ion collisions.

Findings

Enhanced jet momentum reconstruction accuracy.

Improved classification of quenched vs. unquenched jets.

Better identification of jet origins and energy loss.

Abstract

Jets are suppressed and modified in heavy ion collisions, which serve as powerful probes to the properties of the quark-gluon plasma (QGP). Attributed to the abundant information carried by the jet constituents and reconstructed substructures, plenty of interesting applications of machine learning techniques have been made on a jet-by-jet basis to study the jet quenching phenomena. Here we review recent proceedings on this topic including the tasks of reconstructing jet momentum in heavy ion collisions, classifying quenched jets and unquenched jets, identifying jet energy loss, locating the jet creation points as well as distinguishing between quark- and gluon-initiated jets in the QGP. Such jet-by-jet analyses will allow us to have a better handle on the jet reconstruction and selections to investigate the effects of jet modifications and push forward the long-standing goal of jet tomographic probes of the QGP.