NUCLEAR TECHNIQUES, Volume. 46, Issue 8, 080013(2023)

Studies of nuclear β-decay half-lives with Bayesian neural network approach

Weifeng LI, Xiaoyan ZHANG, and Zhongming NIU*
Author Affiliations
  • School of Physics and Optoelectronic and Engineering, Anhui University, Hefei 230601, China
  • show less
    Figures & Tables(7)
    RMS deviations from experimental data σRMS(lgT1/2) calculated by BNN-I3, BNN-I4, and BNN-I5 approaches for training set (a) and validation set (b)
    Nuclear β-decay half-lives and errors for Sn isotopes predicted by BNN-I2, BNN-I3, and BNN-I4 approaches[5]
    Comparison of nuclear β-decay half-lives of Ni, Sn, and Pb isotopic chains predicted by BNN-I4 with the theoretical results of RHB+QRPA, FRDM+QRPA, SHFB+FAM, SHFB+QRPA, and WS4+GT
    Comparison of nuclear β-decay half-lives of N=50, N=82, and N=126 isotonic chains predicted by BNN-I4 with the theoretical results of RHB+QRPA, FRDM+QRPA, SHFB+FAM, SHFB+QRPA, and WS4+GT
    Logarithmic difference distribution on the nuclear chart between the predicted β-decay half-lives by BNN-I4 and experimental data
    Logarithmic difference of β-decay half-lives between the BNN-I4 predictions and theoretical results of WS4+GT
    • Table 1. Input variables, number of hidden layer neurons, activation functions, and output variables of various neural networks

      View table
      View in Article

      Table 1. Input variables, number of hidden layer neurons, activation functions, and output variables of various neural networks

      神经网络

      Neural networks

      输入变量

      Input variables

      隐藏层神经元个数

      Number of hidden layer neurons

      激活函数

      Activation functions

      输出变量

      Output variables

      BNN-I2Z、N30tanhlgT1/2
      BNN-I3Z、N、Qβ24tanhlgT1/2
      BNN-I4Z、N、δ、Qβ20tanhlgT1/2
    Tools

    Get Citation

    Copy Citation Text

    Weifeng LI, Xiaoyan ZHANG, Zhongming NIU. Studies of nuclear β-decay half-lives with Bayesian neural network approach[J]. NUCLEAR TECHNIQUES, 2023, 46(8): 080013

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Research Articles

    Received: Feb. 28, 2023

    Accepted: --

    Published Online: Sep. 19, 2023

    The Author Email:

    DOI:10.11889/j.0253-3219.2023.hjs.46.080013

    Topics