Deep Learning Analysis of Deeply Virtual Exclusive Photoproduction

Abstract

We present a Machine Learning based approach to the cross section and asymmetries for deeply virtual Compton scattering from an unpolarized proton target using both an unpolarized and polarized electron beam. Machine learning methods are needed to study and eventually interpret the outcome of deeply virtual exclusive experiments since these reactions are characterized by a complex final state with a larger number of kinematic variables and observables, exponentially increasing the difficulty of quantitative analyses. Our deep neural network (FemtoNet) uncovers emergent features in the data and learns an accurate approximation of the cross section that outperforms standard baselines. FemtoNet reveals that the predictions in the unpolarized case systematically show a smaller relative median error than the polarized that can be ascribed to the presence of the Bethe Heitler process. It also suggests that the $t$ dependence can be more easily extrapolated than for the other variables, namely the skewness, $\xi$ and four-momentum transfer, $Q^2$. Our approach is fully scalable and will be capable of handling larger data sets as they are released from future experiments.

arXiv comment14 pages, 12 figures

References in corpus (19)

• Adam: A Method for Stochastic Optimization
• Language Models are Few-Shot Learners
• Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning
• Parton distributions from high-precision collider data
• Unraveling hadron structure with generalized parton distributions
• Hard Exclusive Reactions and the Structure of Hadrons
• Deeply virtual electroproduction of photons and mesons on the nucleon : leading order amplitudes and power corrections
• GPD phenomenology and DVCS fitting - Entering the high-precision era
• Generalized parton distributions with helicity flip
• The E00-110 experiment in Jefferson Lab's Hall A: Deeply Virtual Compton Scattering off the Proton at 6 GeV
• Toward solving the sign problem with path optimization method
• Finite-Density Monte Carlo Calculations on Sign-Optimized Manifolds
• Unbiased determination of DVCS Compton Form Factors
• A Glimpse of Gluons through Deeply Virtual Compton Scattering on the Proton
• Extraction of Generalized Parton Distribution Observables from Deeply Virtual Electron Proton Scattering Experiments
• Easy as ${\pi }^o$: On the Interpretation of Recent Electroproduction Results
• Measurement of deeply virtual Compton scattering (DVCS) cross sections with CLAS
• Report from the A.I. For Nuclear Physics Workshop
• Studying Deeply Virtual Compton Scattering with Neural Networks

Cited by in corpus (10)

• 50 Years of Quantum Chromodynamics
• Precision Studies of QCD in the Low Energy Domain of the EIC
• The Present and Future of QCD
• Strong Interaction Physics at the Luminosity Frontier with 22 GeV Electrons at Jefferson Lab
• On Extraction of Twist-Two Compton Form Factors from DVCS Observables Through Harmonic Analysis
• Accessing Compton Form Factors at the Electron Ion Collider in China: An Impact study on $\text{Im}\mathcal{E}$
• Unified and optimal frame choice for generalized parton distributions
• Parametrization of Quark and Gluon Generalized Parton Distributions in a Dynamical Framework
• New framework for extracting GPDs from exclusive photon electroproduction
• Assessing the impact of the electron ion collider in China on Deeply Virtual Compton Scattering