Classifying FRB spectrograms using nonlinear dimensionality reduction techniques

Abstract

Fast radio bursts (FRBs) are mysterious astronomical phenomena, and it is still uncertain whether they consist of multiple types. In this study we use two nonlinear dimensionality reduction algorithms - Uniform Manifold Approximation and Projection (UMAP) and t-distributed stochastic neighbour embedding (t-SNE) - to differentiate repeaters from apparently non-repeaters in FRBs. Based on the first Canadian Hydrogen Intensity Mapping Experiment (CHIME) FRB catalogue, these two methods are applied to standardized parameter data and image data from a sample of 594 sub-bursts and 535 FRBs, respectively. Both methods are able to differentiate repeaters from apparently non-repeaters. The UMAP algorithm using image data produces more accurate results and is a more model-independent method. Our result shows that in general repeater clusters tend to be narrowband, which implies a difference in burst morphology between repeaters and apparently non-repeaters. We also compared our UMAP predictions with the CHIME/FRB discovery of 6 new repeaters, the performance was generally good except for one outlier. Finally, we highlight the need for a larger and more complete sample of FRBs.

arXiv comment10 pages, 9 figures, 4 tables, accepted by MNRAS

References in corpus (30)

• A bright millisecond radio burst of extragalactic origin
• A New Electron Density Model for Estimation of Pulsar and FRB Distances
• A Repeating Fast Radio Burst
• The direct localization of a fast radio burst and its host
• A fast radio burst associated with a Galactic magnetar
• A bright millisecond-duration radio burst from a Galactic magnetar
• A census of baryons in the Universe from localized fast radio bursts
• An extreme magneto-ionic environment associated with the fast radio burst source FRB 121102
• The First CHIME/FRB Fast Radio Burst Catalog
• A repeating fast radio burst source localised to a nearby spiral galaxy
• A single fast radio burst localized to a massive galaxy at cosmological distance
• CHIME/FRB Detection of Eight New Repeating Fast Radio Burst Sources
• Periodic activity from a fast radio burst source
• A fast radio burst localised to a massive galaxy
• A Second Source of Repeating Fast Radio Bursts
• A repeating fast radio burst associated with a persistent radio source
• The low density and magnetization of a massive galaxy halo exposed by a fast radio burst
• A bimodal burst energy distribution of a repeating fast radio burst source
• Possible periodic activity in the repeating FRB 121102
• Nine New Repeating Fast Radio Burst Sources from CHIME/FRB
• An ultra-wide bandwidth (704 to 4032 MHz) receiver for the Parkes radio telescope
• A fast radio burst source at a complex magnetised site in a barred galaxy
• Diverse polarization angle swings from a repeating fast radio burst source
• The prevalence of repeating fast radio bursts
• Faint Repetitions from a Bright Fast Radio Burst Source
• Uncloaking hidden repeating fast radio bursts with unsupervised machine learning
• Implications of CHIME repeating fast radio bursts
• A catalogue of 323 cataclysmic variables from LAMOST DR6
• A semi-supervised Machine Learning search for never-seen Gravitational-Wave sources
• A Possible Subclassification of Fast Radio Bursts

Cited by in corpus (7)

• Exploring the Key Features of Repeating Fast Radio Bursts with Machine Learning
• Unsupervised Machine Learning for Classifying CHIME Fast Radio Bursts and Investigating Empirical Relations
• Positive and unlabelled machine learning reveals new fast radio burst repeater candidates
• Comparative analysis of machine learning techniques for feature selection and classification of Fast Radio Bursts
• Repeating versus Nonrepeating Fast Radio Bursts: A Deep Learning Approach to Morphological Characterization
• A Practical Framework for Estimating the Repetition Likelihood of Fast Radio Bursts from Spectral Morphology
• Unveiling the spectral morphological division of fast radio bursts with CHIME/FRB Catalog 2