Employing Deep Neural Networks for extracting the 3D Structure of Nucleon

Transverse Momentum Dependent distributions (TMDs) of the nucleon can be extracted from processes involving multiple kinematic scales, such as Drell-Yan (DY), Semi-Inclusive Deep Inelastic Scattering (SIDIS), and electron-positron annihilation. In contrast, the cross section from Deeply Virtual Compton Scattering (DVCS) provides access to Compton Form Factors (CFFs), which are convolutions of Generalized Parton Distributions (GPDs) with coefficient functions that are calculable in perturbative Quantum Chromodynamics (pQCD). Because of this convolution structure, GPDs cannot be accessed directly. Deep Neural Networks (DNNs) have emerged as powerful tools for modeling and extracting information from data. We demonstrate two distinct applications of DNNs for the extraction of TMDs and CFFs, highlighting their effectiveness in learning from experimental observables.