Search

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 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