Pulsar Candidate Identification Using Semi-Supervised Generative Adversarial Networks

Abstract

Machine learning methods are increasingly helping astronomers identify new radio pulsars. However, they require a large amount of labelled data, which is time consuming to produce and biased. Here we describe a Semi-Supervised Generative Adversarial Network (SGAN) which achieves better classification performance than the standard supervised algorithms using majority unlabelled datasets. We achieved an accuracy and mean F-Score of 94.9% trained on only 100 labelled candidates and 5000 unlabelled candidates compared to our standard supervised baseline which scored at 81.1% and 82.7% respectively. Our final model trained on a much larger labelled dataset achieved an accuracy and mean F-score value of 99.2% and a recall rate of 99.7%. This technique allows for high quality classification during the early stages of pulsar surveys on new instruments when limited labelled data is available. We open-source our work along with a new pulsar-candidate dataset produced from the High Time Resolution Universe - South Low Latitude Survey. This dataset has the largest number of pulsar detections of any public dataset and we hope it will be a valuable tool for benchmarking future machine learning models.

arXiv commentadded coauthors and one extra plot, main results unchanged, accepted by MNRAS journal

References in corpus (30)

• Adam: A Method for Stochastic Optimization
• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Scikit-learn: Machine Learning in Python
• Array Programming with NumPy
• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Deep Residual Learning for Image Recognition
• Generative Adversarial Text to Image Synthesis
• The Parkes Multibeam Pulsar Survey: I. Observing and Data Analysis Systems, Discovery and Timing of 100 Pulsars
• The High Time Resolution Universe Pulsar Survey I: System configuration and initial discoveries
• Arecibo Pulsar Survey Using ALFA. I. Survey Strategy and First Discoveries
• Generative Adversarial Networks recover features in astrophysical images of galaxies beyond the deconvolution limit
• Fifty Years of Pulsar Candidate Selection: From simple filters to a new principled real-time classification approach
• Good Semi-supervised Learning that Requires a Bad GAN
• The Green Bank Northern Celestial Cap Pulsar Survey - I: Survey Description, Data Analysis, and Initial Results
• A stochastic template placement algorithm for gravitational wave data analysis
• FETCH: A deep-learning based classifier for fast transient classification
• Searching for pulsars using image pattern recognition
• Selection of radio pulsar candidates using artificial neural networks
• CosmoGAN: creating high-fidelity weak lensing convergence maps using Generative Adversarial Networks
• The High Time Resolution Universe Survey VI: An Artificial Neural Network and Timing of 75 Pulsars
• The High Time Resolution Universe Pulsar Survey XII : Galactic plane acceleration search and the discovery of 60 pulsars
• ExoGAN: Retrieving Exoplanetary Atmospheres Using Deep Convolutional Generative Adversarial Networks
• Lense-Thirring frame dragging induced by a fast-rotating white dwarf in a binary pulsar system
• An improved test of the strong equivalence principle with the pulsar in a triple star system
• Separation of pulsar signals from noise with supervised machine learning algorithms
• Investigation of nulling and subpulse drifting properties of PSR J1727$-$2739
• PEACE: Pulsar Evaluation Algorithm for Candidate Extraction -- A software package for post-analysis processing of pulsar survey candidates
• The High Time Resolution Universe Pulsar Survey -- XVI. Discovery and timing of 40 pulsars from the southern Galactic plane
• Interstellar Scintillation observations for PSR B0355+54

Cited by in corpus (9)

• Searching for pulsars associated with polarised point sources using LOFAR: Initial discoveries from the TULIPP project
• Discovery of 37 new pulsars through GPU-accelerated reprocessing of archival data of the Parkes Multibeam Pulsar Survey
• PulsarX: a new pulsar searching package -I. A high performance folding program for pulsar surveys
• Missing for 20 years: MeerKAT re-detects the elusive binary pulsar M30B
• The Southern-sky MWA Rapid Two-metre (SMART) pulsar survey -- II. Survey status, pulsar census, and first pulsar discoveries
• Morphological Classification of Radio Galaxies with wGAN-supported Augmentation
• Analyzing the Galactic pulsar distribution with machine learning
• Classification of the \emph{Fermi}-LAT Blazar Candidates of Uncertain type using eXtreme Gradient Boosting
• Radio Halo Detection in MWA Data using Deep Neural Networks and Generative Data Augmentation