Pulsar candidate selection based on self-normalizing neural networks

Zhi-Wei Kang; Tuo Liu; Jin Liu; Xin Ma; Xiao Chen

doi:10.7498/aps.69.20191582

Journals >Acta Physica Sinica >Volume 69 >Issue 6 >Page 069701-1 > Article

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

Pulsar candidate selection based on self-normalizing neural networks

Zhi-Wei Kang^1、*, 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

DOI: 10.7498/aps.69.20191582 Cite this Article

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 Copy Citation Text

show less

Fig. 1. SELU activation function.

Download full size | View in the Article

Fig. 2. GMO_SNN candidate selection algorithm.

Download full size | View in the Article

Fig. 3. Comparison of the loss function between SNN and ANN.

Download full size | View in the Article

数据集	非脉冲星数	脉冲星数	总样本数
HTRU 1	89996	1196	91192
HTRU 2	16259	1639	17898
LOTAAS 1	4987	66	5053

Table 1.

Pulsar candidate datasets.

脉冲星候选体数据集

编号	特征	编号	特征
1	P	12	轮廓直方图最大值/高斯拟合的最大值
2	DM	13	对轮廓求导后的直方图与轮廓直方图的偏移量
3	S/N	14	$S/N/\sqrt {\left( {P - W} \right)/W} $
4	W	15	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	16	DM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	17	DM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 2.

Feature description.

特征描述

隐藏层数	F1-score/%
2	82.48
4	89.58
8	94.56
9	94.20

Table 3. [in Chinese]

批次大小	F1-score/%	运行时间/s
16	94.87	74
32	94.56	43
64	93.90	23
128	91.05	11

Table 4. [in Chinese]

隐藏层数	F1-score/%
0.1	无法收敛
0.01	94.29
0.001	94.55
0.0001	84.10

Table 5. [in Chinese]

数据集	模型	Accuracy/%	Recall/%	Precison/%	F1-score/%	FPR/%	G-mean/%
HTRU 1	SNN	99.82	92.44	93.45	92.94	0.08	96.11
	GA_SNN	99.85	92.45	95.19	93.80	0.06	96.12
	MO_SNN	99.81	94.23	97.94	96.05	0.05	97.05
	GMO_SNNNNNNNNN	99.85	95.32	98.51	96.89	0.04	97.61

HTRU 2	SNN	98.30	87.73	93.93	90.73	0.59	93.38
	GA_SNN	98.30	88.91	92.86	90.84	0.71	93.96
	MO_SNN	97.89	92.17	95.08	93.60	0.95	95.54
	GMO_SNNNNNNNNN	98.03	92.53	95.58	94.03	0.08	95.78

LOTAAS 1	SNN	99.92	93.75	100.00	96.77	0.08	96.79
	GA_SNN	99.92	100.00	93.33	96.55	0	100.00
	MO_SNN	99.69	100.00	87.10	93.10	0.31	99.84
	GMO_SNN	100.00	100.00	100.00	100.00	0	100.00

Table 6. [in Chinese]

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

Tools

Save the article for my favorites

Paper Information