Acta Physica Sinica, Volume. 69, Issue 6, 069701-1(2020)

Pulsar candidate selection based on self-normalizing neural networks

Zhi-Wei Kang1、*, Tuo Liu1, Jin Liu2, Xin Ma3, and Xiao Chen4
Author Affiliations
  • 1College of Computer Science and Electronic Engineering, Hunan University, Changsha 410082, China
  • 2College of Information Science and Engineering, Wuhan University of Science and Technology, Wuhan 430081, China
  • 3College of Instrument Science and Opto Electronic Engineering, Beihang University, Beijing 100191, China
  • 4Shanghai Institution of Satellite Engineering, Shanghai 200240, China
  • show less
    Figures & Tables(9)
    SELU activation function.
    GMO_SNN candidate selection algorithm.
    Comparison of the loss function between SNN and ANN.
    • Table 1.

      Pulsar candidate datasets.

      脉冲星候选体数据集

      View table
      View in Article

      Table 1.

      Pulsar candidate datasets.

      脉冲星候选体数据集

      数据集非脉冲星数脉冲星数总样本数
      HTRU 189996119691192
      HTRU 216259163917898
      LOTAAS 14987665053
    • Table 2.

      Feature description.

      特征描述

      View table
      View in Article

      Table 2.

      Feature description.

      特征描述

      编号特征编号特征
      1P12轮廓直方图最大值/高斯拟合的最大值
      2DM13对轮廓求导后的直方图与轮廓直方图的偏移量
      3S/N14$S/N/\sqrt {\left( {P - W} \right)/W} $
      4W15拟合 $S/N/\sqrt {\left( {P - W} \right)/W} $
      5用sin曲线拟合脉冲轮廓的卡方值16DM拟合值与DM最优值取余
      6用sin2曲线拟合脉冲轮廓的卡方值 17DM曲线拟合的卡方值
      7高斯拟合脉冲轮廓的卡方值18峰值处对应的所有频段值的均方根
      8高斯拟合脉冲轮廓的半高宽19任意两个频段线性相关度的均值
      9双高斯拟合脉冲轮廓的卡方值20线性相关度的和
      10双高斯拟合脉冲轮廓的平均半高宽21脉冲轮廓的波峰数
      11脉冲轮廓直方图对0的偏移量22脉冲轮廓减去均值后的面积
    • Table 4. [in Chinese]

      View table
      View in Article

      Table 4. [in Chinese]

      批次大小F1-score/%运行时间/s
      1694.8774
      3294.5643
      6493.9023
      12891.0511
    • Table 5. [in Chinese]

      View table
      View in Article

      Table 5. [in Chinese]

      隐藏层数F1-score/%
      0.1无法收敛
      0.0194.29
      0.00194.55
      0.000184.10
    • Table 6. [in Chinese]

      View table
      View in Article

      Table 6. [in Chinese]

      数据集模型Accuracy/%Recall/%Precison/%F1-score/%FPR/%G-mean/%
      HTRU 1SNN99.8292.4493.4592.940.0896.11
      GA_SNN99.8592.4595.1993.800.0696.12
      MO_SNN99.8194.2397.9496.050.0597.05
      GMO_SNNNNNNNNN99.8595.3298.5196.890.0497.61
      HTRU 2SNN98.3087.7393.9390.730.5993.38
      GA_SNN98.3088.9192.8690.840.7193.96
      MO_SNN97.8992.1795.0893.600.9595.54
      GMO_SNNNNNNNNN98.0392.5395.5894.030.0895.78
      LOTAAS 1SNN99.9293.75100.0096.770.0896.79
      GA_SNN99.92100.0093.3396.550100.00
      MO_SNN99.69100.0087.1093.100.3199.84
      GMO_SNN100.00100.00100.00100.000100.00
    Tools

    Get Citation

    Copy Citation Text

    Zhi-Wei Kang, Tuo Liu, Jin Liu, Xin Ma, Xiao Chen. Pulsar candidate selection based on self-normalizing neural networks[J]. Acta Physica Sinica, 2020, 69(6): 069701-1

    Download Citation

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

    Category:

    Received: Oct. 17, 2019

    Accepted: --

    Published Online: Nov. 19, 2020

    The Author Email:

    DOI:10.7498/aps.69.20191582

    Topics