Identifying the nature of the QCD transition in heavy-ion collisions with deep learning
Abstract
In this proceeding, we review our recent work using deep convolutional neural network (CNN) to identify the nature of the QCD transition in a hybrid modeling of heavy-ion collisions. Within this hybrid model, a viscous hydrodynamic model is coupled with a hadronic cascade “after-burner”. As a binary classification setup, we employ two different types of equations of state (EoS) of the hot medium in the hydrodynamic evolution. The resulting final-state pion spectra in the transverse momentum and azimuthal angle plane are fed to the neural network as the input data in order to distinguish different EoS. To probe the effects of the fluctuations in the event-by-event spectra, we explore different scenarios for the input data and make a comparison in a systematic way. We observe a clear hierarchy in the predictive power when the network is fed with the event-by-event, cascade-coarse-grained and event-fine-averaged spectra. The carefully-trained neural network can extract high-level features from pion spectra to identify the nature of the QCD transition in a realistic simulation scenario.
- Authors:
-
- Frankfurt Inst. for Advanced Studies (FIAS), Frankfurt (Germany); Goethe Univ., Frankfurt (Germany); Nanjing Univ. (China); Univ. of Bergen (Norway)
- Frankfurt Inst. for Advanced Studies (FIAS), Frankfurt (Germany)
- Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States); Univ. of California, Berkeley, CA (United States); China Central Normal Univ., Wuhan (China)
- Frankfurt Inst. for Advanced Studies (FIAS), Frankfurt (Germany); Goethe Univ., Frankfurt (Germany)
- Nanjing Univ. (China); Nanjing Proton Source Research and Design Center (China); Anhui Normal Univ., Wuhu (China)
- Frankfurt Inst. for Advanced Studies (FIAS), Frankfurt (Germany); Goethe Univ., Frankfurt (Germany); GSI-Helmholtzzentrum fur Schwerionenforschung, Darmstadt (Germany)
- Publication Date:
- Research Org.:
- Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC); Trond Mohn Foundation; National Natural Science Foundation of China (NSFC); National Major State Basic Research and Development of China; National Science Foundation (NSF)
- OSTI Identifier:
- 1810799
- Grant/Contract Number:
- AC02-05CH11231; BFS2018RED01; 11475085; 11535005; 11690030; 11221504; 2016Y-FE0129300; 2014CB845404; ACI-1550228
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Nuclear Physics. A
- Additional Journal Information:
- Journal Volume: 1005; Journal ID: ISSN 0375-9474
- Publisher:
- Elsevier
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 73 NUCLEAR PHYSICS AND RADIATION PHYSICS; heavy-ion physics; QCD equation of states; hybrid model; deep learning
Citation Formats
Du, Yi-Lun, Zhou, Kai, Steinheimer, Jan, Pang, Long-Gang, Motornenko, Anton, Zong, Hong-Shi, Wang, Xin-Nian, and Stöcker, Horst. Identifying the nature of the QCD transition in heavy-ion collisions with deep learning. United States: N. p., 2020.
Web. doi:10.1016/j.nuclphysa.2020.121891.
Du, Yi-Lun, Zhou, Kai, Steinheimer, Jan, Pang, Long-Gang, Motornenko, Anton, Zong, Hong-Shi, Wang, Xin-Nian, & Stöcker, Horst. Identifying the nature of the QCD transition in heavy-ion collisions with deep learning. United States. https://doi.org/10.1016/j.nuclphysa.2020.121891
Du, Yi-Lun, Zhou, Kai, Steinheimer, Jan, Pang, Long-Gang, Motornenko, Anton, Zong, Hong-Shi, Wang, Xin-Nian, and Stöcker, Horst. Thu .
"Identifying the nature of the QCD transition in heavy-ion collisions with deep learning". United States. https://doi.org/10.1016/j.nuclphysa.2020.121891. https://www.osti.gov/servlets/purl/1810799.
@article{osti_1810799,
title = {Identifying the nature of the QCD transition in heavy-ion collisions with deep learning},
author = {Du, Yi-Lun and Zhou, Kai and Steinheimer, Jan and Pang, Long-Gang and Motornenko, Anton and Zong, Hong-Shi and Wang, Xin-Nian and Stöcker, Horst},
abstractNote = {In this proceeding, we review our recent work using deep convolutional neural network (CNN) to identify the nature of the QCD transition in a hybrid modeling of heavy-ion collisions. Within this hybrid model, a viscous hydrodynamic model is coupled with a hadronic cascade “after-burner”. As a binary classification setup, we employ two different types of equations of state (EoS) of the hot medium in the hydrodynamic evolution. The resulting final-state pion spectra in the transverse momentum and azimuthal angle plane are fed to the neural network as the input data in order to distinguish different EoS. To probe the effects of the fluctuations in the event-by-event spectra, we explore different scenarios for the input data and make a comparison in a systematic way. We observe a clear hierarchy in the predictive power when the network is fed with the event-by-event, cascade-coarse-grained and event-fine-averaged spectra. The carefully-trained neural network can extract high-level features from pion spectra to identify the nature of the QCD transition in a realistic simulation scenario.},
doi = {10.1016/j.nuclphysa.2020.121891},
journal = {Nuclear Physics. A},
number = ,
volume = 1005,
place = {United States},
year = {Thu Dec 10 00:00:00 EST 2020},
month = {Thu Dec 10 00:00:00 EST 2020}
}
Works referenced in this record:
Onset of classical QCD dynamics in relativistic heavy ion collisions
journal, May 2005
- Kharzeev, Dmitri; Levin, Eugene; Nardi, Marzia
- Physical Review C, Vol. 71, Issue 5
An equation-of-state-meter of quantum chromodynamics transition from deep learning
journal, January 2018
- Pang, Long-Gang; Zhou, Kai; Su, Nan
- Nature Communications, Vol. 9, Issue 1
The iEBE-VISHNU code package for relativistic heavy-ion collisions
dataset, January 2019
- Shen, Chun
- Mendeley
Constraining the Equation of State of Superhadronic Matter from Heavy-Ion Collisions
journal, May 2015
- Pratt, Scott; Sangaline, Evan; Sorensen, Paul
- Physical Review Letters, Vol. 114, Issue 20
QCD equation of state and hadron resonance gas
journal, June 2010
- Huovinen, Pasi; Petreczky, Péter
- Nuclear Physics A, Vol. 837, Issue 1-2
Microscopic models for ultrarelativistic heavy ion collisions
journal, January 1998
- Bass, S.
- Progress in Particle and Nuclear Physics, Vol. 41
GLISSANDO: GLauber Initial-State Simulation AND mOre…
journal, January 2009
- Broniowski, Wojciech; Rybczyński, Maciej; Bożek, Piotr
- Computer Physics Communications, Vol. 180, Issue 1
Color glass condensate at the LHC: hadron multiplicities in pp, pA and AA collisions
journal, January 2005
- Kharzeev, Dmitri; Levin, Eugene; Nardi, Marzia
- Nuclear Physics A, Vol. 747, Issue 2-4
Applying Bayesian parameter estimation to relativistic heavy-ion collisions: Simultaneous characterization of the initial state and quark-gluon plasma medium
journal, August 2016
- Bernhard, Jonah E.; Moreland, J. Scott; Bass, Steffen A.
- Physical Review C, Vol. 94, Issue 2
The iEBE-VISHNU code package for relativistic heavy-ion collisions
dataset, January 2019
- Shen, Chun
- Mendeley
Hydrodynamical description of 200 A GeV/ c S+Au collisions: Hadron and electromagnetic spectra
journal, January 1997
- Sollfrank, Josef; Huovinen, Pasi; Kataja, Markku
- Physical Review C, Vol. 55, Issue 1
Relativistic hadron-hadron collisions in the ultra-relativistic quantum molecular dynamics model
journal, September 1999
- Bleicher, M.; Zabrodin, E.; Spieles, C.
- Journal of Physics G: Nuclear and Particle Physics, Vol. 25, Issue 9
The iEBE-VISHNU code package for relativistic heavy-ion collisions
journal, February 2016
- Shen, Chun; Qiu, Zhi; Song, Huichao
- Computer Physics Communications, Vol. 199