Determination of impact parameter in high-energy heavy-ion collisions via deep learning
arXiv:2112.03824 · doi:10.1088/1674-1137/ac6490
Abstract
In this study, Au+Au collisions with the impact parameter of fm at GeV are simulated by the AMPT model to provide the preliminary final-state information. After transforming these information into appropriate input data (the energy spectra of final-state charged hadrons), we construct a deep neural network (DNN) and a convolutional neural network (CNN) to connect final-state observables with impact parameters. The results show that both the DNN and CNN can reconstruct the impact parameters with a mean absolute error about fm with CNN behaving slightly better. Then, we test the neural networks for different beam energies and pseudorapidity ranges in this task. It turns out that these two models work well for both low and high energies. But when making test for a larger pseudorapidity window, we observe that the CNN shows higher prediction accuracy than the DNN. With the method of Grad-CAM, we shed light on the `attention' mechanism of the CNN model.
References in corpus (25)
- Solving the Quantum Many-Body Problem with Artificial Neural Networks
- Glauber Modeling in High Energy Nuclear Collisions
- A Multi-Phase Transport Model for Relativistic Heavy Ion Collisions
- Collective flow and viscosity in relativistic heavy-ion collisions
- Searching for Exotic Particles in High-Energy Physics with Deep Learning
- Formation of superdense hadronic matter in high energy heavy-ion collisions
- ZPC 1.0.1: a parton cascade for ultrarelativistic heavy ion collisions
- Search for the QCD Critical Point with Fluctuations of Conserved Quantities in Relativistic Heavy-Ion Collisions at RHIC : An Overview
- Centrality determination of Pb-Pb collisions at = 2.76 TeV with ALICE
- Event-by-event shape and flow fluctuations of relativistic heavy-ion collision fireballs
- Electromagnetic fields and anomalous transports in heavy-ion collisions --- A pedagogical review
- Pseudorapidity distributions of charged particles from Au+Au collisions at the maximum RHIC energy, Sqrt(s_NN) = 200 GeV
- Deep learning and the Schrödinger equation
- Novel quantum phenomena induced by strong magnetic fields in heavy-ion collisions
- Anomalous chiral transports and spin polarization in heavy-ion collisions
- Identifying the nature of the QCD transition in relativistic collision of heavy nuclei with deep learning
- Impact parameter dependence of the azimuthal asymmetry in lepton pair production in heavy ion collisions
- Neural Networks for Impact Parameter Determination
- Nuclear liquid-gas phase transition with machine learning
- A fast centrality-meter for heavy-ion collisions at the CBM experiment
- Machine-learning-based identification for initial clustering structure in relativistic heavy-ion collisions
- Application of artificial intelligence in the determination of impact parameter in heavy-ion collisions at intermediate energies
- Application of machine learning in the determination of impact parameter in the Sn+Sn system
- Estimation of Impact Parameter and Transverse Spherocity in heavy-ion collisions at the LHC energies using Machine Learning
- Applications of deep learning to relativistic hydrodynamics