The Symposium on Frontier Innovation and Development in Nuclear Science and Technology & the Third Plenary Editorial Board Meeting of the Two Journals, organized by the journals Nuclear Techniques and Nuclear Science and Techniques, was recently held at Fudan University (Jiangwan Campus) in Shanghai.

　　A paper titled “Studies on Several Problems in Nuclear Physics Using Machine Learning,” with Professor Li Qingfeng (Huzhou University) as the corresponding author and Gao Zepeng (a master’s student jointly trained with Huzhou University, now a Ph.D. candidate at Sun Yat-sen University) as the first author, won the 2024 Nuclear Techniques Outstanding Paper Award. In addition, Professor Wang Yongjia (Huzhou University) was named Outstanding Young Editorial Board Member of Nuclear Techniques in recognition of his contributions to enhancing the journal’s academic quality and international impact.

　　In recent years, machine learning has shown great promise in nuclear physics, enabling researchers to uncover deep physical patterns from complex, large-scale datasets. The award-winning study evaluates convolutional neural networks (CNNs) and LightGBM across three frontier tasks—refining nuclear-mass models, reconstructing heavy-ion collision impact parameters, and extracting the slope parameter of the nuclear symmetry energy (L)—and reports the following results:

　　(1) The root mean square deviation of the nuclear mass formula was optimized to below 0.2 MeV, maintaining good accuracy in extrapolation tasks.

　　(2) For 1 GeV/nucleon Au+Au collisions, the impact parameter reconstruction accuracy reached Δb < 0.1 fm, and for 0.27 GeV/nucleon Sn+Sn collisions, it was Δb ≈ 0.3–0.4 fm.

　　(3) The classification accuracy for five different symmetry energies was 58% (proton spectra) and 72% (neutron spectra), respectively.

　　(4) When using 30 feature quantities, the mean absolute error between the predicted and true values of the symmetry energy slope parameter was less than 30 MeV.

　　These findings provide new ideas and methods for tackling key problems in nuclear physics. The research was supported by the National Natural Science Foundation of China and the National Key R&D Program of China.

　　About the Journals. Nuclear Techniques is sponsored by the Shanghai Institute of Applied Physics, Chinese Academy of Sciences, and the Chinese Nuclear Society. Launched in 1978, it is indexed by EI Compendex and the Peking University Core Journals database, and is listed as Tier T1 in the Chinese Nuclear Society’s 2024 Journal Classification for the Nuclear Field. Nuclear Science and Techniques, launched in 1990, is the English journal of the Chinese Nuclear Society (CNS). It is indexed by Scopus, EI, SCI-E, and INSPIRE, and is ranked Q1 in the “Physics & Astronomy” major category of the Chinese Academy of Sciences (CAS) Journal Ranking, as well as Q1 in the “Nuclear Science & Technology” and “Physics, Nuclear” subcategories.

　　Link to the paper:

https://www.hjs.sinap.ac.cn/zh/article/doi/10.11889/j.0253-3219.2023.hjs.46.080009/