In this work, we looked into using artificial neural networks to analyze heavy ion collision data on an event-by-event basis. Using simulated data from a microchannel plate detector (MCP) [1], we concentrated on resolving the impact parameter evaluation and collision vertex coordinate estimation problem for possible application in NICA collider experiments [2]. Based on the QGSM event generator [3] data, our research shows that this kind of method can be used to estimate collision parameters fairly accurately, resulting in average error of about
[4, 5, 6]. And this method extracts collision information from raw detector data, namely from particles spatial distributions and their time-of-flight distributions.

However, the model of event generator that was used to create the dataset has a significant impact on the results of ANNs. Different outcomes were obtained when the experiments were repeated using data from alternative generator models [7, 8]. In this work we discuss how to use ANNs to create model-independent algorithms despite model dependencies in data. Furthermore, we demonstrate the possible application of our results, based on the fact that detector parameters that yield the best reconstruction of the event parameters are independent of the Monte-Carlo model used to simulate events.

The authors acknowledge Saint-Petersburg State University for a research project 95413904.

References:
[1] A.A.Baldin, G.A. Feofilov, P. Har'yuzov, and F.F. Valiev,
// Nucl. Instrum. Meth.A 2020, V.958, P.162154. https://doi.org/10.1016/j.nima.2019.04.108
[2] https://nica.jinr.ru/
[3] Amelin N. S., Gudima K. K., Toneev V. D. Ultrarelativistic nucleus-nucleus collisions within a dynamical model of independent quark - gluon strings // Sov. J. Nucl. Phys. 1990. V. 51(6), P. 1730-1743
[4] K.A. Galaktionov, V.A. Roudnev, and F.F. Valiev, Neural network approach to impact parameter estimation in high-energy collisions using the microchannel plate detector data,
// Moscow University Physics Bulletin 2023, V. 78, P. S52-S58
[5] Galaktionov K.A., Roudnev V.A., Valiev F.F. Artificial Neural Networks Application in Estimating the Impact Parameter in Heavy Ion Collisions Using the Microchannel Plate Detector Data: Physics of Atomic Nuclei.
//Phys. At. Nucl. 2023 V.86(6), P.1426-1432. https://doi.org/10.1134/S1063778823060248
[6] Galaktionov, K., Roudnev, V., Valiev, F., Application of Neural Networks for Event-by-Event Evaluation of the Impact Parameter,
// Physics of Particles and Nuclei 2023 ,V. 54, P. 446-448
[7] Werner, Klaus and Liu, Fu-Ming and Pierog, Tanguy Parton ladder splitting and the rapidity dependence of transverse momentum spectra in deuteron-gold collisions at the BNL Relativistic Heavy Ion Collider
// Physical Review C 2006, V. 74
[8] Aichelin, J. and Bratkovskaya, E. and Le Fèvre, A. and Kireyeu, V. and Kolesnikov, V. and Leifels, Y. and Voronyuk, V. and Coci, G. Parton-hadron-quantum-molecular dynamics: A novel microscopic n-body transport approach for heavy-ion collisions, dynamical cluster formation, and hypernuclei production
// Physical Review C 2020, V. 101