Machine learning spectral functions in lattice QCD

Abstract

We study the inverse problem of reconstructing spectral functions from Euclidean correlation functions via machine learning. We propose a novel neural network, SVAE, which is based on the variational autoencoder (VAE) and can be naturally applied to the inverse problem. The prominent feature of the SVAE is that a Shannon-Jaynes entropy term having the ground truth values of spectral functions as prior information is included in the loss function to be minimized. We train the network with general spectral functions produced from a Gaussian mixture model. As a test, we use correlators generated from four different types of physically motivated spectral functions made of one resonance peak, a continuum term and perturbative spectral function obtained using non-relativistic QCD. From the mock data test we find that the SVAE in most cases is comparable to the maximum entropy method (MEM) in the quality of reconstructing spectral functions and even outperforms the MEM in the case where the spectral function has sharp peaks with insufficient number of data points in the correlator. By applying to temporal correlation functions of charmonium in the pseudoscalar channel obtained in the quenched lattice QCD at 0.75 $T_c$ on $128^3\times 96$ lattices and $1.5$ $T_c$ on $128^3\times 48$ lattices, we find that the resonance peak of ${\eta }_c$ extracted from both the SVAE and MEM has a substantial dependence on the number of points in the temporal direction ($N_{\tau }$) adopted in the lattice simulation and $N_{\tau }$ larger than 48 is needed to resolve the fate of ${\eta }_c$ at 1.5 $T_c$.

25 pages, 14 figures. Investigations on the dependences of output spectral functions on the noise model of mock correlators, and detailed derivation of formulae for the output spectral function are added

References in corpus (35)

• Machine learning and the physical sciences
• $J/\psi$ and ${\eta }_c$ in the Deconfined Plasma from Lattice QCD
• Thermal dilepton rate and electrical conductivity: An analysis of vector current correlation functions in quenched lattice QCD
• J/$\psi$ suppression at forward rapidity in Pb-Pb collisions at $\sqrt{s_{\mathrm{NN}}}$ = 2.76 TeV
• Nuclear mass predictions based on Bayesian neural network approach with pairing and shell effects
• Sequential charmonium dissociation
• A novel Bayesian approach to spectral function reconstruction
• Charmonium properties in hot quenched lattice QCD
• Static quark-antiquark potential in the quark-gluon plasma from lattice QCD
• Continuum and lattice meson spectral functions at nonzero momentum and high temperature
• Estimation of Thermodynamic Observables in Lattice Field Theories with Deep Generative Models
• Meson Screening Masses in (2+1)-Flavor QCD
• Suppression of Upsilon Production in d+Au and Au+Au Collisions at sqrt(s_NN) = 200 GeV
• Regressive and generative neural networks for scalar field theory
• Heavy-flavor production and medium properties in high-energy nuclear collisions - What next?
• Artificial Neural Network Approach to the Analytic Continuation Problem
• Deep Learning and AdS/CFT
• Infinite temperature limit of meson spectral functions calculated on the lattice
• Dielectron Mass Spectra from Au+Au Collisions at $\sqrt{s_{\mathrm{NN}}}$ = 200 GeV
• Thermal dilepton rates and electrical conductivity of the QGP from the lattice
• Analytic continuation via domain-knowledge free machine learning
• Deep Learning and Holographic QCD
• Spectral Reconstruction with Deep Neural Networks
• Machine learning action parameters in lattice quantum chromodynamics
• Measurement of ${\rm Y}$(1S+2S+3S) production in $p$$+$$p$ and Au$+$Au collisions at $\sqrt{s_{{}_{NN}}}=200$ GeV
• Lattice constraints on the thermal photon rate
• Suppression of excited ${\rm Y}$ states relative to the ground state in PbPb collisions at $\sqrt{s_{\mathrm{NN}}}$ = 5.02 TeV
• The pion quasiparticle in the low-temperature phase of QCD
• Machine-learning physics from unphysics: Finding deconfinement temperature in lattice Yang-Mills theories from outside the scaling window
• New developments in lattice QCD on equilibrium physics and phase diagram
• Machine Learning Estimators for Lattice QCD Observables
• Stochastic reconstructions of spectral functions: Application to lattice QCD
• Towards reduction of autocorrelation in HMC by machine learning
• Bryan's Maximum Entropy Method -- diagnosis of a flawed argument and its remedy
• Comment on "Heavy Quarkonium in Extreme Conditions"

Cited by in corpus (3)

• Machine Learning in Nuclear Physics
• Neural network reconstruction of the dense matter equation of state from neutron star observables
• Bayesian inference of real-time dynamics from lattice QCD