Deep learning predicted elliptic flow of identified particles in heavy-ion collisions at the RHIC and LHC energies
Neelkamal Mallick, Suraj Prasad, A. N. Mishra, R. Sahoo, G. G. Barnaföldi
Abstract
Recent developments of a deep learning feed-forward network for estimating elliptic flow (${v}_{2}$) coefficients in heavy-ion collisions have shown the prediction power of this technique. The success of the model is mainly the estimation of ${v}_{2}$ from final-state particle kinematic information and learning the centrality and transverse momentum (${p}_{\mathrm{T}}$) dependence of ${v}_{2}$. The deep learning model is trained with Pb-Pb collisions at $\sqrt{{s}_{\mathrm{NN}}}=5.02\text{ }\text{ }\mathrm{TeV}$ minimum bias events simulated with a multiphase transport model. We extend this work to estimate ${v}_{2}$ for light-flavor identified particles such as ${\ensuremath{\pi}}^{\ifmmode\pm\else\textpm\fi{}}$, ${\mathrm{K}}^{\ifmmode\pm\else\textpm\fi{}}$, and $\mathrm{p}+\overline{\mathrm{p}}$ in heavy-ion collisions at RHIC and LHC energies. The number-of-constituent-quark scaling is also shown. The evolution of the ${p}_{\mathrm{T}}$-crossing point of ${v}_{2}({p}_{\mathrm{T}})$, depicting a change in baryon-meson elliptic flow at intermediate ${p}_{\mathrm{T}}$, is studied for various collision systems and energies. The model is further evaluated by training it for different ${p}_{\mathrm{T}}$ regions. These results are compared with the available experimental data wherever possible.