A vision of hadronic physics

Abstract

Abstract. We present a vision for the next decade of hadron physics in which the central question being addressed is how one might win new physical insight into the way hadronic systems work. The topics addressed include the relevance of model building, the role of spontaneously broken chiral symmetry, spectroscopy, form factors and physics in the deep inelastic regime.

Applications of neural networks in hadron physics

“…The hard photon contribution of the TPE is induced by the hadronic proton structure. Many efforts have been made to calculate this contribution based on the theoretical and phenomenological models (see references [6][7][8][9][10][11][12][13][14][15][16][17][18] in [11]). However, for larger Q 2 values the predictions of TPE are model-dependent.…”

“…The prediction is plotted together with the error bounds ( σ 1 error due the distribution of weights) see equations (3.15) and (3.16) of [10]. The horizontal lines denote the mean value (weighted by the evidence) see (14), while the shaded areas denote the σ 1 uncertainty given by Δ + − R , calculated as described above. Notice that the best model predictions agree well with the new data from [26], which were not included in the analysis.…”

“…In the following sections we shall introduce the Bayesian framework (BF) for feedforward neural networks and as an instructive example of application we will present results of our studies of the proton structure, E-M FFs and the two-photon exchange (TPE) effect [13,11]. Investigation of the proton FFs and related observables (proton radius, two photon exchange effect) is an important topic of the hadron physics [14]. In this paper we discuss the statistical features of the framework concentrating our attention on the quantitative model comparison and the estimate of the systematic uncertainties due to the choice of initial model assumptions.…”

“…Each single point denotes the value of R +/− calculated for one particular model, in this case the error bounds are related with the weights uncertainty. The solid lines show the mean values calculated from(14). The shaded area denotes 1σ uncertainty due to the model-dependence of the parametrizations.…”