Finding signatures of the nuclear symmetry energy in heavy-ion collisions with deep learning

Abstract

A deep convolutional neural network (CNN) is developed to study symmetry energy $E_{\mathrm{sym}}(\rho )$ effects by learning the mapping between the symmetry energy and the two-dimensional (transverse momentum and rapidity) distributions of protons and neutrons in heavy-ion collisions. Supervised training is performed with labelled data-set from the ultrarelativistic quantum molecular dynamics (UrQMD) model simulation. It is found that, by using proton spectra on event-by-event basis as input, the accuracy for classifying the soft and stiff $E_{\mathrm{sym}}(\rho )$ is about 60% due to large event-by-event fluctuations, while by setting event-summed proton spectra as input, the classification accuracy increases to 98%. The accuracy for 5-label (5 different $E_{\mathrm{sym}}(\rho )$) classification task are about 58% and 72% by using proton and neutron spectra, respectively. For the regression task, the mean absolute error (MAE) which measures the average magnitude of the absolute differences between the predicted and actual $L$ (the slope parameter of $E_{\mathrm{sym}}(\rho )$) are about 20.4 and 14.8 MeV by using proton and neutron spectra, respectively. Fingerprints of the density-dependent nuclear symmetry energy on the transverse momentum and rapidity distributions of protons and neutrons can be identified by convolutional neural network algorithm.

arXiv comment6 pages, 7 figures

References in corpus (26)

• Machine learning and the physical sciences
• Microscopic Models for Ultrarelativistic Heavy Ion Collisions
• Relativistic Hadron-Hadron Collisions in the Ultra-Relativistic Quantum Molecular Dynamics Model (UrQMD)
• Recent Progress and New Challenges in Isospin Physics with Heavy-Ion Reactions
• Equations of state for supernovae and compact stars
• Constraints on the symmetry energy and neutron skins from experiments and theory
• Skyrme Interaction and Nuclear Matter Constraints
• Modelling the many-body dynamics of heavy ion collisions: Present status and future perspective
• The Nuclear Symmetry Energy
• Nuclear Equation of State from ground and collective excited state properties of nuclei
• Progress in Constraining Nuclear Symmetry Energy Using Neutron Star Observables Since GW170817
• Towards Understanding Astrophysical Effects of Nuclear Symmetry Energy
• Nucleon Effective Masses in Neutron-Rich Matter
• Nuclear charge radii: Density functional theory meets Bayesian neural networks
• Shannon Information Entropy in Heavy-ion Collisions
• Dynamics of clusters and fragments in heavy-ion collisions
• Nuclear dynamics and particle production near threshold energies in heavy-ion collisions
• Transport approaches for the Description of Intermediate-Energy Heavy-Ion Collisions
• Identifying the nature of the QCD transition in relativistic collision of heavy nuclei with deep learning
• Collision dynamics at medium and relativistic energies
• Neural Networks for Impact Parameter Determination
• Unsupervised Outlier Detection in Heavy-Ion Collisions
• A fast centrality-meter for heavy-ion collisions at the CBM experiment
• CSHINE for studies of HBT correlation in Heavy Ion Reactions
• Determining the temperature in heavy-ion collisions with multiplicity distribution
• Constraining isovector nuclear interactions with giant resonances within a Bayesian approach

Cited by in corpus (2)

• Phase Transition Study meets Machine Learning
• Bayesian Data Fusion of Imperfect Fission Yields for Augmented Evaluations