Dec 7 – 10, 2026
IPEN
America/Sao_Paulo timezone

Exploring Jet Quenching in Simulated Heavy-Ion Collisions with Machine Learning Models

Dec 9, 2026, 2:00 PM
15m
152 - Auditorium Prof. Dr. Afonso Rodrigues de Aquino (IPEN)

152 - Auditorium Prof. Dr. Afonso Rodrigues de Aquino

IPEN

Oral Relativistic heavy ion collisions Oral HIC

Speaker

Leonardo Lima Da Silva (Universidade de Sao Paulo (USP) (BR))

Description

The Monte Carlo event generator JEWEL (Jet Evolution With Energy Loss) provides a theoretical framework to simulate parton energy loss in a dense medium such as the Quark–Gluon Plasma (QGP). Comparisons between theory and experiment rely on global observables, including the nuclear modification factor $R_{AA}$ and the dijet momentum imbalance $x_J$. In this work, we take a different approach by adopting a jet-by-jet analysis to study quenched jets with supervised machine learning architectures.
For this purpose, we employ the code JEWEL with both the default configuration, which employs Glauber initial conditions coupled to Bjorken hydrodynamic model, and a more complex medium description, combining T$_\mathrm{R}$ENTo initial conditions with the (2+1)D hydrodynamic evolution of v-USPhydro.
Each jet is represented as a sequential trajectory along the Cambridge/Aachen declustering tree, characterized by four substructure observables ($z$, $\Delta R$, $k_\perp$, $m_{inv}$) at each splitting, which we analyze using two supervised machine learning architectures: Long Short-Term Memory (LSTM) networks and Transformers.
Our results show that the machine learning models chosen in this work are effective for the binary classification problem of distinguishing quenched jets on a jet-by-jet basis, with accuracies and AUC scores greater than 95\%. In addition, the trained models are sensitive to medium properties that are hidden in traditional global observables, demonstrating the potential of ML for probing the Quark-Gluon Plasma.
Finally, we briefly discuss ongoing extensions of this analysis aimed at moving beyond the binary classification of PbPb and pp jets. In particular, we explore strategies to characterize different degrees of jet quenching on a jet-by-jet basis, using matched vacuum–medium jet comparisons. These developments seek to connect the information learned by the models with physically interpretable quantities, providing a bridge between supervised classification and a more differential description of jet energy loss.

Keywords Jet quenching, Quark-Gluon Plasma, Machine learning
Proceedings Publication Intent Yes
Student Award Eligibility yes
Financial Support Request no

Author

Leonardo Lima Da Silva (Universidade de Sao Paulo (USP) (BR))

Co-author

Marcelo Gameiro Munhoz (Universidade de Sao Paulo (BR))

Presentation materials

There are no materials yet.