Machine-Learning Prediction for Quasi-PDF Matrix Elements

Abstract

There have been rapid developments in the direct calculation in lattice QCD (LQCD) of the Bjorken-$x$ dependence of hadron structure through large-momentum effective theory (LaMET). LaMET overcomes the previous limitation of LQCD to moments (that is, integrals over Bjorken-$x$) of hadron structure, allowing LQCD to directly provide the kinematic regions where the experimental values are least known. LaMET requires large-momentum hadron states to minimize its systematics and allow us to reach small-$x$ reliably. This means that very fine lattice spacing to minimize lattice artifacts at order $(P_{z}a{)}^n$ will become crucial for next-generation LaMET-like structure calculations. Furthermore, such calculations require operators with long Wilson-link displacements (in finer lattice units), increasing the communication costs relative to that of the propagator inversion. In this work, we explore whether machine-learning (ML) algorithms can make correlator predictions to reduce the computational cost of these LQCD calculations. We consider two algorithms, gradient-boosting decision tree and linear models, applied to LaMET data, the matrix elements needed to determine the kaon and ${\eta }_s$ unpolarized parton distribution functions (PDFs), meson distribution amplitude (DA), and the nucleon gluon PDF. We find that both algorithms can reliably predict the target observables with different fit quality and systematic errors. The predictions from smaller displacement $z$ to larger ones work better than those for momentum $p$ due to the higher correlation among the data.

arXiv comment20 pages, 22 figures

References in corpus (36)

• Scikit-learn: Machine Learning in Python
• QCD Factorization for $B\to \pi \pi$ Decays: Strong Phases and CP Violation in the Heavy Quark Limit
• QCD Factorization in B -> pi K, pi pi Decays and Extraction of Wolfenstein Parameters
• The Chroma Software System for Lattice QCD
• Parton Physics on Euclidean Lattice
• Flavor Symmetry and the Static Potential with Hypercubic Blocking
• Highly Improved Staggered Quarks on the Lattice, with Applications to Charm Physics
• Domain wall QCD with physical quark masses
• Quasi-PDFs, momentum distributions and pseudo-PDFs
• Lattice QCD ensembles with four flavors of highly improved staggered quarks
• Parton distributions and lattice QCD calculations: a community white paper
• Nucleon Helicity and Transversity Parton Distributions from Lattice QCD
• Flavor Structure of the Nucleon Sea from Lattice QCD
• Novel quark smearing for hadrons with high momenta in lattice QCD
• A Lattice Calculation of Parton Distributions
• A complete non-perturbative renormalization prescription for quasi-PDFs
• Pion and Kaon Structure at the Electron-Ion Collider
• Perturbative Renormalization of quasi-PDFs
• Pion Distribution Amplitude from Lattice QCD
• Nonperturbative renormalization of nonlocal quark bilinears for quasi-PDFs on the lattice using an auxiliary field
• Parton Distribution Function with Non-perturbative Renormalization from Lattice QCD
• Pion Valence Structure from Ioffe Time Pseudo-Distributions
• Updated Lattice Results for Parton Distributions
• First direct lattice-QCD calculation of the $x$-dependence of the pion parton distribution function
• Reconstructing parton distribution functions from Ioffe time data: from Bayesian methods to Neural Networks
• Gluon Quasi-PDF From Lattice QCD
• Improved Parton Distribution Functions at Physical Pion Mass
• Parton Distribution Functions from Ioffe time pseudo-distributions
• Hadrons and Nuclei
• Physics of Gluon Helicity Contribution to Proton Spin
• Gluon Helicity Delta G from a Universality Class of Operators on a Lattice
• Justifying the Naive Partonic Sum Rule for Proton Spin
• Gaussian-weighted Parton Quasi-distribution
• Machine Learning Estimators for Lattice QCD Observables
• Kaon Distribution Amplitude from Lattice QCD and the Flavor SU(3) Symmetry
• Symmetry Properties of Nonlocal Quark Bilinear Operators on a Lattice

Cited by in corpus (20)

• Machine Learning in Nuclear Physics
• Parton distributions and lattice QCD calculations: toward 3D structure
• Exploring QCD matter in extreme conditions with Machine Learning
• Valence-Quark Distribution of the Kaon and Pion from Lattice QCD
• Nucleon Tomography and Generalized Parton Distribution at Physical Pion Mass from Lattice QCD
• Pion and Kaon Distribution Amplitudes in the Continuum Limit
• Nuclear matrix elements from lattice QCD for electroweak and beyond-Standard-Model processes
• $B$-meson light-cone distribution amplitude from the Euclidean quantity
• One-loop matching for the twist-3 parton distribution $g_T(x)$
• Nucleon Gluon Distribution Function from 2+1+1-Flavor Lattice QCD
• Learning Lattice Quantum Field Theories with Equivariant Continuous Flows
• One-Loop Hybrid Renormalization Matching Kernels for Quasi-Parton Distributions
• Renormalon Effects in Quasi Parton Distributions
• Matching the B-meson quasidistribution amplitude in the RI/MOM scheme
• Chiral Perturbation for Large Momentum Effective Field Theory
• Lossy compression of statistical data using quantum annealer
• Mapping Parton Distributions of Hadrons with Lattice QCD
• Report from the A.I. For Nuclear Physics Workshop
• Large-Momentum Effective Theory's Asymptotic Extrapolation vs the Inverse Problem
• One-loop Matching Factors for Singlet Quasi-Parton Distribution Functions in the Hybrid-Ratio Scheme