Pulsar candidate selection using ensemble networks for FAST drift-scan survey

Wang, HongFeng; Zhu, Weiwei; Guo, Ping; Li, Di; Feng, Sibo; Yin, Qian; Miao, Chenchen; Tao, Zhen-Zhao; Pan, Zhichen; Wang, Pei; Zheng, Xin; Deng, Xiaoya; Liu, Zhijie; Xie, Xiaoyao; Yu, Xuhong; You, Shijun; Zhang, Hui

doi:10.1007/s11433-018-9388-3

Cited by 37 publications

(23 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Due to the deployment of deep learning approaches in diverse fields for classification and their high accuracy, several deep learning-based models have been adopted for pulsar detection and classification. For example, the authors used a convolutional neural network (CNN) in the PCIS algorithm from the ResNet model for pulsar detection in [15]. On the GBNCC dataset, the proposed system achieved 96% accuracy.…”

Section: Related Workmentioning

confidence: 99%

Predicting Pulsars from Imbalanced Dataset with Hybrid Resampling Approach

Lee

Rustam

Aljedaani

et al. 2021

Advances in Astronomy

View full text Add to dashboard Cite

Pulsar stars, usually neutron stars, are spherical and compact objects containing a large quantity of mass. Each pulsar star possesses a magnetic field and emits a slightly different pattern of electromagnetic radiation which is used to identify the potential candidates for a real pulsar star. Pulsar stars are considered an important cosmic phenomenon, and scientists use them to study nuclear physics, gravitational waves, and collisions between black holes. Defining the process of automatic detection of pulsar stars can accelerate the study of pulsar stars by scientists. This study contrives an accurate and efficient approach for true pulsar detection using supervised machine learning. For experiments, the high time-resolution (HTRU2) dataset is used in this study. To resolve the data imbalance problem and overcome model overfitting, a hybrid resampling approach is presented in this study. Experiments are performed with imbalanced and balanced datasets using well-known machine learning algorithms. Results demonstrate that the proposed hybrid resampling approach proves highly influential to avoid model overfitting and increase the prediction accuracy. With the proposed hybrid resampling approach, the extra tree classifier achieves a 0.993 accuracy score for true pulsar star prediction.

show abstract

Section: Related Workmentioning

confidence: 99%

Predicting Pulsars from Imbalanced Dataset with Hybrid Resampling Approach

Lee

Rustam

Aljedaani

et al. 2021

Advances in Astronomy

View full text Add to dashboard Cite

show abstract

“…The automated systems to classify folded candidates have been using various machine learning techniques such as decision trees, support vector machines and neural networks (see e.g. Zhu et al 2014;Tan et al 2018;Wang et al 2019) Folded pulsar-candidate plots try to summarize the observational characteristics of the candidate. Information that has been discarded during the initial search process like the temporal and spectral persistence of the pulsar signal are shown in such typical diagnostic plots (Tan et al 2018, Fig 1).…”

Section: Machine Learning In Pulsar Astronomymentioning

confidence: 99%

Detecting Pulsars with Neural Networks: A Proof of Concept

Künkel,

Thomas,

Verbiest

2021

Preprint

View full text Add to dashboard Cite

Pulsar searches are computationally demanding efforts to discover dispersed periodic signals in time-and frequency-resolved data from radio telescopes. The complexity and computational expense of simultaneously determining the frequency-dependent delay (dispersion) and the periodicity of the signal is further exacerbated by the presence of various types of radiofrequency interference (RFI) and observing-system effects. New observing systems with wider bandwidths, higher bit rates and greater overall sensitivity (also to RFI) further enhance these challenges. We present a novel approach to the analysis of pulsar search data. Specifically, we present a neural-network-based pipeline that efficiently suppresses a wide range of RFI signals and instrumental instabilities and furthermore corrects for (a priori unknown) interstellar dispersion. After initial training of the network, our analysis can be run in real time on a standard desktop computer with a commonly available, consumer-grade GPU. We complement our neural network with standard algorithms for periodicity searches. In particular with the Fast Fourier Transform (FFT) and the Fast Folding Algorithm (FFA) and demonstrate that with these straightforward extensions, our method is capable of identifying even faint pulsars, while maintaining an extremely low number of false positives. We furthermore apply our analysis to a subset of the PALFA survey and demonstrate that in most cases the automated dispersion removal of our network produces a time series of similar quality as dedispersing using the actual dispersion measure of the pulsar in question. On our test data we are able to make predictions whether a pulsar is present in the data or not 200 times faster than real time.

show abstract

“…FAST自2016年建成以来进行了密集的调试, 并于2020年1月完成了国家验收. 在FAST调试期间, 通过实测确定了FAST的基本性能 [9] , 并已经取得了一系列成果 [10][11][12][13][14][15] . 这证明了FAST The Five-hundred-meter Aperture Spherical Radio Telescope (FAST)-a large-aperture radio telescope constructed in a karst depression-is completing its commissioning.…”

Section: 讨论和总结unclassified