اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدDeep Learning Analysis of Deeply Virtual Exclusive Photoproduction
81
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
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.
قيم البحث
اقرأ أيضاً
The goal of the comprehensive program in Deeply Virtual Exclusive Scattering at Jefferson Laboratory is to create transverse spatial images of quarks and gluons as a function of their longitudinal momentum fraction in the proton, the neutron, and in
nuclei. These functions are the Generalized Parton Distributions (GPDs) of the target nucleus. Cross section measurements of the Deeply Virtual Compton Scattering (DVCS) reaction {ep -> ep gamma} in Hall A support the QCD factorization of the scattering amplitude for Q^2 > 2 GeV^2. Quasi-free neutron-DVCS measurements on the Deuteron indicate sensitivity to the quark angular momentum sum rule. Fully exclusive H(e,e pgamma) measurements have been made in a wide kinematic range in CLAS with polarized beam, and with both unpolarized and longitudinally polarized targets. Existing models are qualitatively consistent with the JLab data, but there is a clear need for less constrained models. Deeply virtual vector meson production is studied in CLAS. The 12 GeV upgrade will be essential for these channels. The {rho} and {omega} channels reactions offer the prospect of flavor sensitivity to the quark GPDs, while the {phi}-production channel is dominated by the gluon distribution.
We study tensor meson photoproduction outside of the resonance region, at beam energies of few GeVs. We build a model based on Regge theory that includes the leading vector and axial exchanges. We consider two determinations of the unknown helicity c
ouplings, and fit to the recent a2 photoproduction data from CLAS. Both choices give a similar description of the a2 cross section, but result in different predictions for the parity asymmetries and the f2 photoproduction cross section. We conclude that new measurements of f2 photoproduction in the forward region are needed to pin down the correct production mechanism. We also extend our predictions to the 8.5 GeV beam energy, where current experiments are running.
In this talk it is reported on analyses of l p -> l pi+ n and pi- p -> l+ l- n within the handbag approach. It is argued that recent measurements of hard pion production performed by HERMES and CLAS clearly indicate the occurrence of strong contribut
ions from transversely polarized photons. The gamma*T -> pi transitions are described by the transversity GPDs accompanied by twist-3 pion wave functions. The experiments also require strong contributions from the pion pole which can be modeled as a classical one-pion exchange. With these extensions the handbag approach leads to results on cross sections and spin asymmetries in fair agreement with experiment. This approach is also used for an estimate of the partial cross sections for the exclusive Drell-Yan process.
We study the use of deep learning techniques to reconstruct the kinematics of the deep inelastic scattering (DIS) process in electron-proton collisions. In particular, we use simulated data from the ZEUS experiment at the HERA accelerator facility, a
nd train deep neural networks to reconstruct the kinematic variables $Q^2$ and $x$. Our approach is based on the information used in the classical construction methods, the measurements of the scattered lepton, and the hadronic final state in the detector, but is enhanced through correlations and patterns revealed with the simulated data sets. We show that, with the appropriate selection of a training set, the neural networks sufficiently surpass all classical reconstruction methods on most of the kinematic range considered. Rapid access to large samples of simulated data and the ability of neural networks to effectively extract information from large data sets, both suggest that deep learning techniques to reconstruct DIS kinematics can serve as a rigorous method to combine and outperform the classical reconstruction methods.
Diffractive deeply virtual Compton scattering (DiDVCS) is the process $gamma^*(- Q^2) + N rightarrow rho^0 + gamma^* (Q^2)+ N$, where N is a nucleon or light nucleus, in the kinematical regime of large rapidity gap between the $rho^0$ and the final p
hoton-nucleus system, and in the generalized Bjorken regime where both photon virtualities $Q^2$ and $ Q^2$ are large. We show that this process has the unique virtue of combining the large diffractive cross sections at high energy with the tomographic ability of deeply virtual Compton scattering to scrutinize the quark and gluon content of nucleons and light nuclei. Its study at an electron-ion collider would enlighten the internal structure of hadrons.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
