Speaker
Description
This work focuses on the challenge of estimating multiple parameters in heavy-ion collisions using data-driven techniques [1-3]. A significant difficulty arises from the dependence of algorithm accuracy on the choice of event generator models such as QGSM [4], EPOS [5], and PHQMD [6], which introduce biases affecting parameter reconstruction.
We evaluated several approaches to mitigate these biases, starting with classical methods like principal component analysis (PCA) and naive training on combined datasets from different generators. Then, we explored neural network-based methods, including deep reconstruction networks [7], which demonstrated superior robustness and accuracy.
The best-performing models achieved accuracy comparable to those trained on single-generator data, while improving over naive mixed-dataset training. These findings suggest that advanced neural network techniques can help overcome generator-induced biases, supporting the development of generalized algorithms for reliable estimation of heavy-ion collision parameters.
The authors acknowledge Saint-Petersburg State University for a research project 103821868
[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] K.A. Galaktionov, V.A. Roudnev, and F.F. Valiev, Moscow Univ. Phys. Bull. 78 (2023) Suppl 1, S52-S58
[4] Amelin N. S., Gudima K. K., Toneev V. D., Sov. J. Nucl. Phys. 1990. V. 51(6), P. 1730-1743
[5] Werner, Klaus and Liu, Fu-Ming and Pierog, Physical Review C 2006, V. 74
[6] Aichelin, J. and Bratkovskaya, E. and Le Fèvre, A. and Kireyeu, V. and Kolesnikov, V. and Leifels, Y. and Voronyuk, V. and Coci, G., Physical Review C 2020, V. 101
[7] Wang, M. and Deng, W. Neurocomputing, 2018, V. 312, P 135-153
