Search

Quantitatively connecting properties of parton distribution functions (PDFs, or parton densities) to the theoretical assumptions made within the QCD analyses which produce them has been a longstanding problem in HEP phenomenology. To confront this challenge, we introduce an ML-based explainability framework, $\texttt{XAI4PDF}$, to classify PDFs by parton flavor or underlying theoretical model using ResNet-like neural networks (NNs). By leveraging the differentiable nature of ResNet models, this approach deploys guided backpropagation to dissect relevant features of fitted PDFs, identifying x-dependent signatures of PDFs important to the ML model classifications. By applying our framework, we are able to sort PDFs according to the analysis which produced them while constructing quantitative, human-readable maps locating the x regions most affected by the internal theory assumptions going into each analysis. This technique expands the toolkit available to PDF analysis and adjacent particle phenomenology while pointing to promising generalizations., Comment: 25 pages, 6 figures, 1 table; for link to associated public code, see https://github.com/ML4HEP-Theory/XAI4PDF

We develop a new methodology for extracting Compton form factors (CFFs) in from deeply virtual exclusive reactions such as the unpolarized DVCS cross section using a specialized inverse problem solver, a variational autoencoder inverse mapper (VAIM). The VAIM-CFF framework not only allows us access to a fitted solution set possibly containing multiple solutions in the extraction of all 8 CFFs from a single cross section measurement, but also accesses the lost information contained in the forward mapping from CFFs to cross section. We investigate various assumptions and their effects on the predicted CFFs such as cross section organization, number of extracted CFFs, use of uncertainty quantification technique, and inclusion of prior physics information. We then use dimensionality reduction techniques such as principal component analysis to visualize the missing physics information tracked in the latent space of the VAIM framework. Through re-framing the extraction of CFFs as an inverse problem, we gain access to fundamental properties of the problem not comprehensible in standard fitting methodologies: exploring the limits of the information encoded in deeply virtual exclusive experiments., Comment: 17 pages, 15 figures

Representing the parton distribution functions (PDFs) of the proton and other hadrons through flexible, high-fidelity parametrizations has been a long-standing goal of particle physics phenomenology. This is particularly true since the chosen parametrization methodology can play an influential role in the ultimate PDF uncertainties as extracted in QCD global analyses; these, in turn, are often determinative of the reach of experiments at the LHC and other facilities to non-standard physics, including at large $x$, where parametrization effects can be significant. In this study, we explore a series of encoder-decoder machine-learning (ML) models with various neural-network topologies as efficient means of reconstructing PDFs from meaningful information stored in an interpretable latent space. Given recent effort to pioneer synergies between QCD analyses and lattice-gauge calculations, we formulate a latent representation based on the behavior of PDFs in Mellin space, i.e., their integrated moments, and test the ability of various models to decode PDFs from this information faithfully. We introduce a numerical package, $\texttt{PDFdecoder}$, which implements several encoder-decoder models to reconstruct PDFs with high fidelity and use this tool to explore strengths and pitfalls of neural-network approaches to PDF parametrization. We additionally dissect patterns of learned correlations between encoded Mellin moments and reconstructed PDFs which suggest opportunities for further improvements to ML-based approaches to PDF parametrizations and uncertainty quantification., Comment: 22 pages, 11 figures; updated link to public code, https://github.com/ML4HEP-Theory/PDFDecoder

We outline the physics opportunities provided by the Electron Ion Collider (EIC). These include the study of the parton structure of the nucleon and nuclei, the onset of gluon saturation, the production of jets and heavy flavor, hadron spectroscopy and tests of fundamental symmetries. We review the present status and future challenges in EIC theory that have to be addressed in order to realize this ambitious and impactful physics program, including how to engage a diverse and inclusive workforce. In order to address these many-fold challenges, we propose a coordinated effort involving theory groups with differing expertise is needed. We discuss the scientific goals and scope of such an EIC Theory Alliance., Comment: 44 pages, ReVTeX, White Paper on EIC Theory Alliance

We develop a framework to establish benchmarks for machine learning and deep neural networks analyses of exclusive scattering cross sections (FemtoNet). Within this framework we present an extraction of Compton form factors for deeply virtual Compton scattering from an unpolarized proton target. Critical to this effort is a study of the effects of physics constraint built into machine learning (ML) algorithms. We use the Bethe-Heitler process, which is the QED radiative background to deeply virtual Compton scattering, to test our ML models and, in particular, their ability to generalize information extracted from data. We then use these techniques on the full cross section and compare the results to analytic model calculations. We propose a quantification technique, the random targets method, to begin understanding the separation of aleatoric and epistemic uncertainties as they are manifest in exclusive scattering analyses. We propose a set of both physics driven and machine learning based benchmarks providing a stepping stone towards applying explainable machine learning techniques with controllable uncertainties in a wide range of deeply virtual exclusive processes., Comment: 16 pages, 12 figures

We study the deeply virtual Compton scattering process with both twist-two and twist-three Compton form factors and present our cross-sections formulas with all polarization configurations. While the twist-three contributions are generally assumed to be negligible in the literature due to the kinematical suppression, we compare them with the twist-two ones at typical JLab 6 GeV and 12 GeV kinematics as well as EIC kinematics and show their kinematical suppression explicitly, justifying the leading-twist approximation made in the literature. In addition, we also estimate the twist-three Compton form factors using Wandzura-Wilczek relations and inputs of twist-two generalized parton distributions based on a reggeized spectator model. With those estimated Compton form factors, we analyze the kinematical behavior of twist-two and twist-three cross-sections in a wide range of kinematics, and discuss the optimal regions for separating the leading-twist effects from the higher-twist ones., Comment: 53 pages, 13 figures and 1 table

We present a parametrization of the chiral even generalized parton distributions, $H$, $E$, $\widetilde{H}$, $\widetilde{E}$, for the quark, antiquark and gluon, in the perturbative QCD-parton framework. Parametric analytic forms are given as a function of two equivalent sets of variables $x,\xi,t$ (symmetric frame) and $X,\zeta,t$ (asymmetric frame), at an initial scale, $Q_o^2$. In the $X>\zeta$ region a convenient and flexible form is obtained as the product of a Regge term $\propto X^{-\alpha + \alpha' t}$, describing the low $X$ behavior, times a spectator model-based functional form depending on various mass parameters; the behavior at $X<\zeta$, is determined using the generalized parton distributions symmetry and polynomiality properties. The parameters are constrained using data on the flavor separated nucleon electromagnetic elastic form factors, the axial and pseudoscalar nucleon form factors, and the parton distribution functions from both the deep inelastic unpolarized and polarized nucleon structure functions. For the gluon distributions we use, in particular, constraints provided by recent lattice QCD moments calculations. The parametrization's kinematical range of validity is: $0.0001 \leq X \leq 0.85$, $0.01 \leq \zeta \leq 0.85$, $0 \leq -t \leq 1$ GeV$^2$, $2 \leq Q^2 \leq 100$ GeV$^2$. With the simultaneous description of the quark, anti-quark and gluon sectors, this parametrization represents a first tool enabling a global QCD analysis of deeply virtual exclusive experiments., Comment: 28 pages, 19 figures

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., Comment: 14 pages, 12 figures

We use a generalization of the Rosenbluth separation method for a model independent simultaneous extraction of the Compton Form Factors ${\cal H}$ and ${\cal E}$, from virtual Compton scattering data on an unpolarized target. A precise evaluation of ${\cal H}$ and ${\cal E}$, enabled by the proposed method, is the first step towards pinning down the the distribution of angular momentum inside the proton., Comment: 5 pages, 4 figures

Using the helicity amplitudes formalism, we study deeply virtual exclusive electron photo-production off an unpolarized nucleon target, $ep \rightarrow e' p' \gamma$, through a range of kinematics both in the fixed target setting with initial electron energies of 6 GeV, 11 GeV and 24 GeV, and for an electron ion collider. We reformulate the cross section bringing to the forefront the defining features of the $ep \rightarrow e'p'\gamma$ process, where the observables are expressed as bilinear products of the independent helicity amplitudes which completely describe it in terms of the electric, magnetic and axial currents of the nucleon. %This structure is reflected in different ways in the Deeply virtual Compton scattering and Bethe-Heitler parts of the cross section. % These contributions are checked against the Fourier harmonics-based formalism which has provided so far the underlying mathematical framework to study Deeply virtual Compton scattering and related experiments. Using theoretical model calculations of the twist-two generalized parton distributions, $ H$, $E$, $\widetilde{H}$ and $\widetilde{E}$, we uncover large discrepancies between the harmonic series and our proposed framework. Most importantly, these numerical differences appear in the intermediate $Q^2$ range which represents a sweet spot for extracting generalized parton distributions from data. % We provide a framework that is ideal, on one side, to study and compare the different conventions that can be used to describe the leading order contribution to DVCS in QCD, while on the other, it facilitates a quantitative extraction of physically meaningful information from experiment through traceable and controllable approximations in the intermediate $Q^2$ region., Comment: 22 pages, 15 figures

We provide the general expression of the cross section for exclusive deeply virtual photon electroproduction from a spin 1/2 target using current parameterizations of the off-forward correlation function in a nucleon for different beam and target polarization configurations up to twist three accuracy. All contributions to the cross section including deeply virtual Compton scattering, the Bethe-Heitler process, and their interference, are described within a helicity amplitude based framework which is also relativistically covariant and readily applicable to both the laboratory frame and in a collider kinematic setting. Our formalism renders a clear physical interpretation of the various components of the cross section by making a connection with the known characteristic structure of the electron scattering coincidence reactions. In particular, we focus on the total angular momentum, $J_z$, and on the orbital angular momentum, $L_z$. On one side, we uncover an avenue to a precise extraction of $J_z$, given by the combination of generalized parton distributions, $H+E$, through a generalization of the Rosenbluth separation method used in elastic electron proton scattering. On the other, we single out for the first time, the twist three angular modulations of the cross section that are sensitive to $L_z$. The proposed generalized Rosenbluth technique adds an important constraint for mapping the 3D structure of the nucleon., Comment: 62 pages, 7 figures, 3 tables