Early-fusion Based Pulsar Identification with Smart Under-sampling. (arXiv:2107.03071v3 [astro-ph.IM] UPDATED)

Early-fusion Based Pulsar Identification with Smart Under-sampling. (arXiv:2107.03071v3 [astro-ph.IM] UPDATED)
<a href="http://arxiv.org/find/astro-ph/1/au:+Zhang_S/0/1/0/all/0/1">ShiChuan Zhang</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Kong_X/0/1/0/all/0/1">XiangCong Kong</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Zhou_Y/0/1/0/all/0/1">YueYing Zhou</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Chen_L/0/1/0/all/0/1">LingYao Chen</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Zheng_X/0/1/0/all/0/1">XiaoYing Zheng</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Xu_C/0/1/0/all/0/1">Chun-Ling Xu</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Lao_B/0/1/0/all/0/1">Bao-Qiang Lao</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+An_T/0/1/0/all/0/1">Tao An</a>

The discovery of pulsars is of great significance in the field of physics and
astronomy. As the astronomical equipment produces a large amount of pulsar
data, an algorithm for automatically identifying pulsars becomes urgent. We
propose a deep learning framework for pulsar recognition. In response to the
extreme imbalance between positive and negative examples and the hard negative
sample issue presented in the HTRU Medlat Training Data,there are two coping
strategies in our framework: the smart under-sampling and the improved loss
function. We also apply the early-fusion strategy to integrate features
obtained from different attributes before classification to improve the
performance. To our best knowledge,this is the first study that integrates
these strategies and techniques together in pulsar recognition. The experiment
results show that our framework outperforms previous works with the respect to
either the training time or F1 score. We can not only speed up the training
time by 10X compared with the state-of-the-art work, but also get a competitive
result in terms of F1 score.

The discovery of pulsars is of great significance in the field of physics and
astronomy. As the astronomical equipment produces a large amount of pulsar
data, an algorithm for automatically identifying pulsars becomes urgent. We
propose a deep learning framework for pulsar recognition. In response to the
extreme imbalance between positive and negative examples and the hard negative
sample issue presented in the HTRU Medlat Training Data,there are two coping
strategies in our framework: the smart under-sampling and the improved loss
function. We also apply the early-fusion strategy to integrate features
obtained from different attributes before classification to improve the
performance. To our best knowledge,this is the first study that integrates
these strategies and techniques together in pulsar recognition. The experiment
results show that our framework outperforms previous works with the respect to
either the training time or F1 score. We can not only speed up the training
time by 10X compared with the state-of-the-art work, but also get a competitive
result in terms of F1 score.

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

Comments are closed.