Machine-learning-based identification for initial clustering structure in relativistic heavy-ion collisions

TL;DR

This paper demonstrates that a Bayesian convolutional neural network can accurately classify initial nuclear configurations in heavy-ion collisions, achieving high accuracy and robustness, thus opening new avenues for analyzing collision data.

Contribution

The study introduces a machine learning approach using BCNN to distinguish initial clustered and non-clustered nuclear configurations in heavy-ion collisions, with high accuracy and potential for real data application.

Findings

Classification accuracy reaches 95% for specific collision events.

Predicted deviations on mixed samples are within 5%.

Confidence threshold improves prediction on mixed datasets.

Abstract

$\alpha$-clustering structure is a significant topic in light nuclei. A Bayesian convolutional neural network (BCNN) is applied to classify initial non-clustered and clustered configurations, namely Woods-Saxon distribution and three-$\alpha$ triangular (four-$\alpha$ tetrahedral) structure for $^{12}$C ($^{16}$O), from heavy-ion collision events generated within a multi-phase transport (AMPT) model. Azimuthal angle and transverse momentum distributions of charged pions are taken as inputs to train the classifier. On multiple-event basis, the overall classification accuracy can reach $95\%$ for $^{12}$C/$^{16}$O + $^{197}$Au events at $\sqrt{S_{NN}} =$ 200 GeV. With proper constructions of samples, the predicted deviations on mixed samples with different proportions of both configurations could be within $5\%$. In addition, setting a simple confidence threshold can further improve the predictions on the mixed dataset. Our results indicate promising and extensive possibilities of application of machine-learning-based techniques to real data and some other problems in physics of heavy-ion collisions.