Pulsar Candidate Selection Using Ensemble Networks for FAST Drift-Scan Survey. (arXiv:1903.06383v1 [astro-ph.IM])
<a href="http://arxiv.org/find/astro-ph/1/au:+Wang_H/0/1/0/all/0/1">Hongfeng Wang</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Zhu_W/0/1/0/all/0/1">Weiwei Zhu</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Guo_P/0/1/0/all/0/1">Ping Guo</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Li_D/0/1/0/all/0/1">Di Li</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Feng_S/0/1/0/all/0/1">Sibo Feng</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Yin_Q/0/1/0/all/0/1">Qian Yin</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Miao_C/0/1/0/all/0/1">Chenchen Miao</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Tao_Z/0/1/0/all/0/1">Zhenzhao Tao</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Pan_Z/0/1/0/all/0/1">Zhichen Pan</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Wang_P/0/1/0/all/0/1">Pei Wang</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Zheng_X/0/1/0/all/0/1">Xin Zheng</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Liu_X/0/1/0/all/0/1">Xiaodan Deng Zhijie Liu</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Xie_X/0/1/0/all/0/1">Xiaoyao Xie</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Yu_X/0/1/0/all/0/1">Xuhong Yu</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+You_S/0/1/0/all/0/1">Shanping You</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Zhang_H/0/1/0/all/0/1">Hui Zhang</a> (FAST Collaboration)

The Commensal Radio Astronomy Five-hundred-meter Aperture Spherical radio
Telescope (FAST) Survey (CRAFTS) utilizes the novel drift-scan commensal survey
mode of FAST and can generate billions of pulsar candidate signals. The human
experts are not likely to thoroughly examine these signals, and various machine
sorting methods are used to aid the classification of the FAST candidates. In
this study, we propose a new ensemble classification system for pulsar
candidates. This system denotes the further development of the pulsar
image-based classification system (PICS), which was used in the Arecibo
Telescope pulsar survey, and has been retrained and customized for the FAST
drift-scan survey. In this study, we designed a residual network model
comprising 15 layers to replace the convolutional neural networks (CNNs) in
PICS. The results of this study demonstrate that the new model can sort >96% of
real pulsars to belong the top 1% of all candidates and classify >1.6 million
candidates per day using a dual–GPU and 24–core computer. This increased
speed and efficiency can help to facilitate real-time or quasi-real-time
processing of the pulsar-search data stream obtained from CRAFTS. In addition,
we have published the labeled FAST data used in this study online, which can
aid in the development of new deep learning techniques for performing pulsar
searches.

The Commensal Radio Astronomy Five-hundred-meter Aperture Spherical radio
Telescope (FAST) Survey (CRAFTS) utilizes the novel drift-scan commensal survey
mode of FAST and can generate billions of pulsar candidate signals. The human
experts are not likely to thoroughly examine these signals, and various machine
sorting methods are used to aid the classification of the FAST candidates. In
this study, we propose a new ensemble classification system for pulsar
candidates. This system denotes the further development of the pulsar
image-based classification system (PICS), which was used in the Arecibo
Telescope pulsar survey, and has been retrained and customized for the FAST
drift-scan survey. In this study, we designed a residual network model
comprising 15 layers to replace the convolutional neural networks (CNNs) in
PICS. The results of this study demonstrate that the new model can sort >96% of
real pulsars to belong the top 1% of all candidates and classify >1.6 million
candidates per day using a dual–GPU and 24–core computer. This increased
speed and efficiency can help to facilitate real-time or quasi-real-time
processing of the pulsar-search data stream obtained from CRAFTS. In addition,
we have published the labeled FAST data used in this study online, which can
aid in the development of new deep learning techniques for performing pulsar
searches.

http://arxiv.org/icons/sfx.gif