Machine learning simulates collective flow and nuclear stopping in heavy-ion collisions at intermediate energies

Table 1. The parameters of the nuclear matter properties
View table
View in Article
Table 1. The parameters of the nuclear matter properties
参数Parameters 先验分布Prior
$K_{0}$ [180, 380]
$m^{*}$ [0.6, 0.95]
$F$ [0.6, 1.0]

Table 2. Selection of observables

View table

View in Article

Table 2. Selection of observables

观测量Observable		约化碰撞参数Reduced impact parameter b₀	横向动量Transverse 4-velocities u_t0
Feature 1	$v_{11}$	b₀<0.25	u_t0>0.8
Feature 2	$v_{11}$	0.25<b₀<0.45
Feature 3	$v_{20}$	b₀<0.25
Feature 4	$v_{20}$	0.25<b₀<0.45
Feature 5	$v a r t l$	b₀<0.15	None

Table 3. Transport model emulator calibration results
View table
View in Article
Table 3. Transport model emulator calibration results
机器学习Machine learning 训练集Training set R² 测试集Test set R²
高斯过程Gaussian process 0.95 0.95
多任务神经网络Multi-task neural network 0.93 0.93
随机森林Random forest 0.98 0.85

Tools

Get Citation

Copy Citation Text

Guojun WEI, Yongjia WANG, Qingfeng LI, Fuhu LIU. Machine learning simulates collective flow and nuclear stopping in heavy-ion collisions at intermediate energies[J]. NUCLEAR TECHNIQUES, 2025, 48(5): 050005

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites