Efficient emulation of relativistic heavy ion collisions with transfer learning

Abstract

Measurements from the Large Hadron Collider (LHC) and the Relativistic Heavy Ion Collider (RHIC) can be used to study the properties of quark-gluon plasma. Systematic constraints on these properties must combine measurements from different collision systems and methodically account for experimental and theoretical uncertainties. Such studies require a vast number of costly numerical simulations. While computationally inexpensive surrogate models ("emulators") can be used to efficiently approximate the predictions of heavy ion simulations across a broad range of model parameters, training a reliable emulator remains a computationally expensive task. We use transfer learning to map the parameter dependencies of one model emulator onto another, leveraging similarities between different simulations of heavy ion collisions. By limiting the need for large numbers of simulations to only one of the emulators, this technique reduces the numerical cost of comprehensive uncertainty quantification when studying multiple collision systems and exploring different models.

arXiv comment15 pages, 6 figures, journal article

References in corpus (32)

• Bayes in the sky: Bayesian inference and model selection in cosmology
• New Forms of QCD Matter Discovered at RHIC
• Elliptic and triangular flow in event-by-event (3+1)D viscous hydrodynamics
• Hydrodynamic Modeling of Heavy-Ion Collisions
• 3+1D hydrodynamic simulation of relativistic heavy-ion collisions
• Applying Bayesian parameter estimation to relativistic heavy-ion collisions: simultaneous characterization of the initial state and quark-gluon plasma medium
• Alternative ansatz to wounded nucleon and binary collision scaling in high-energy nuclear collisions
• Fully integrated transport approach to heavy ion reactions with an intermediate hydrodynamic stage
• Particle production and equilibrium properties within a new hadron transport approach for heavy-ion collisions
• The production of photons in relativistic heavy-ion collisions
• Multi-system Bayesian constraints on the transport coefficients of QCD matter
• Space-time evolution of bulk QCD matter
• Dynamics of Hot Bulk QCD Matter: from the Quark-Gluon Plasma to Hadronic Freeze-Out
• Bayesian analysis of heavy ion collisions with the heavy ion computational framework Trajectum
• Phenomenological constraints on the transport properties of QCD matter with data-driven model averaging
• Viscous QCD matter in a hybrid hydrodynamic+Boltzmann approach
• A transverse momentum differential global analysis of Heavy Ion Collisions
• Determining Fundamental Properties of Matter Created in Ultrarelativistic Heavy-Ion Collisions
• Estimating initial state and quark-gluon plasma medium properties using a hybrid model with nucleon substructure calibrated to $p$-Pb and Pb-Pb collisions at $\sqrt{s_{\mathrm{NN}}}=5.02$ TeV
• Mass ordering of differential elliptic flow and its violation for phi mesons
• Get on the BAND Wagon: A Bayesian Framework for Quantifying Model Uncertainties in Nuclear Dynamics
• Effects of bulk viscosity and hadronic rescattering in heavy ion collisions at RHIC and LHC
• Quantifying properties of hot and dense QCD matter through systematic model-to-data comparison
• The viscosity of quark-gluon plasma at RHIC and the LHC
• Spectra and elliptic flow for identified hadrons in 2.76 A TeV Pb+Pb collisions
• Coupling Relativistic Viscous Hydrodynamics to Boltzmann Descriptions
• Free-streaming approximation in early dynamics of relativistic heavy-ion collisions
• Towards a Deeper Understanding of How Experiments Constrain the Underlying Physics of Heavy-Ion Collisions
• Constraining the initial state granularity with bulk observables in Au+Au collisions at $\sqrt{s_{\mathrm{NN}}}=200$ GeV
• Emulation of physical processes with Emukit
• A hybrid model approach for strange and multi-strange hadrons in 2.76 A TeV Pb+Pb collisions
• Function-on-function kriging, with applications to 3D printing of aortic tissues

Cited by in corpus (7)

• Dense Nuclear Matter Equation of State from Heavy-Ion Collisions
• Exploring QCD matter in extreme conditions with Machine Learning
• The Present and Future of QCD
• Bayesian quantification of strongly-interacting matter with color glass condensate initial conditions
• Early-times Yang-Mills dynamics and the characterization of strongly interacting matter with statistical learning
• Bayesian analysis of (3+1)D relativistic nuclear dynamics with the RHIC beam energy scan data
• Applications of emulation and Bayesian methods in heavy-ion physics