EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep learning approach for predicting multiple observables in Au + Au collisions at energies available at the BNL Relativistic Heavy Ion Collider

Jun-Qi Tao, Xiang Fan, Yang Liu, Yu Sha, Kai Zhou, Hua Zheng, Ben-Wei Zhang

Physical Review C · 29.06.2026 · 113 (6)

DOI: 10.1103/grkg-jm2z ↗

Kurz gefasst

Fachartikel in der Zeitschrift Physical Review C, online erschienen am 29.06.2026. Verfasst von Jun-Qi Tao und 6 weiteren Autor:innen. Die Arbeit nennt 61 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Ministry of Science and Technology of the People's Republic of China, China Scholarship Council.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We present a data-driven deep learning framework for predicting multiple bulk observables in Au + Au collisions at energies available at the BNL Relativistic Heavy Ion Collider (RHIC). A single neural network is trained exclusively on experimental measurements of charged-particle pseudorapidity density distributions, transverse momentum spectra, and elliptic flow coefficients over a broad range of collision energies and centralities, without using simulation outputs as training targets. The network architecture is inspired by the stages of a heavy-ion collision, from the quark-gluon plasma to chemical and kinetic freeze-out, and employs locally connected hidden layers and a structured input design that encodes basic geometric and kinematic features of the system. We demonstrate that these physics-motivated choices significantly improve test performance compared to purely fully connected baselines. The trained model is then used to predict the above observables at energies available at the RHIC, and the results are further cross-checked against the energy dependence of the total charged-particle multiplicity per participant pair, as well as against a CLVisc hydrodynamic calculation with T R ENT o initial conditions. Our findings indicate that such physics-guided neural networks can provide useful data-driven interpolation tools for RHIC observables in regions not currently covered by the available measurements and can support further phenomenological studies.

Förderung

  • National Natural Science Foundation of China
  • Ministry of Science and Technology of the People's Republic of China
  • China Scholarship Council
  • China Postdoctoral Science Foundation

Zitieren

APA

Tao, J., Fan, X., Liu, Y., Sha, Y., Zhou, K., Zheng, H., & Zhang, B. (2026). Deep learning approach for predicting multiple observables in Au + Au collisions at energies available at the BNL Relativistic Heavy Ion Collider. Physical Review C, 113(6). https://doi.org/10.1103/grkg-jm2z

Vancouver

Tao J, Fan X, Liu Y, Sha Y, Zhou K, Zheng H, et al. Deep learning approach for predicting multiple observables in Au + Au collisions at energies available at the BNL Relativistic Heavy Ion Collider. Physical Review C. 2026;113(6). doi:10.1103/grkg-jm2z

BibTeX
@article{Tao2026,
  author = {Tao, Jun-Qi and Fan, Xiang and Liu, Yang and Sha, Yu and Zhou, Kai and Zheng, Hua and Zhang, Ben-Wei},
  title = {Deep learning approach for predicting multiple observables in Au + Au collisions at energies available at the BNL Relativistic Heavy Ion Collider},
  journal = {Physical Review C},
  year = {2026},
  volume = {113},
  number = {6},
  doi = {10.1103/grkg-jm2z}
}
Kostenlos starten →