Sifting the debris: Patterns in the SNR population with unsupervised ML methods

Abstract

Supernova remnants (SNRs) carry vast amounts of mechanical and radiative energy that heavily influence the structural, dynamical, and chemical evolution of galaxies. To this day, more than 300 SNRs have been discovered in the Milky Way, exhibiting a wide variety of observational features. However, existing classification schemes are mainly based on their radio morphology. In this work, we introduce a novel unsupervised deep learning pipeline to analyse a representative subsample of the Galactic SNR population ($\sim$ 50% of the total) with the aim of finding a connection between their multi-wavelength features and their physical properties. The pipeline involves two stages: (1) a representation learning stage, consisting of a convolutional autoencoder that feeds on imagery from infrared and radio continuum surveys (WISE 22$\mu$m, Hi-GAL 70 $\mu$m and SMGPS 30 cm) and produces a compact representation in a lower-dimensionality latent space; and (2) a clustering stage that seeks meaningful clusters in the latent space that can be linked to the physical properties of the SNRs and their surroundings. Our results suggest that this approach, when combined with an intermediate uniform manifold approximation and projection (UMAP) reprojection of the autoencoded embeddings into a more clusterable manifold, enables us to find reliable clusters. Despite a large number of sources being classified as outliers, most clusters relate to the presence of distinctive features, such as the distribution of infrared emission, the presence of radio shells and pulsar wind nebulae, and the existence of dust filaments.

Accepted in A&A. 17 pages, 11 figures

References in corpus (25)

• The Wide-field Infrared Survey Explorer (WISE): Mission Description and Initial On-orbit Performance
• Herschel Space Observatory - An ESA facility for far-infrared and submillimetre astronomy
• The Photodetector Array Camera and Spectrometer (PACS) on the Herschel Space Observatory
• The Herschel-SPIRE instrument and its in-flight performance
• Clouds, filaments and protostars: the Herschel Hi-GAL Milky Way
• Rotation-invariant convolutional neural networks for galaxy morphology prediction
• EMU: Evolutionary Map of the Universe
• A Census of High-Energy Observations of Galactic Supernova Remnants
• CMU DeepLens: Deep Learning For Automatic Image-based Galaxy-Galaxy Strong Lens Finding
• Automated Transient Identification in the Dark Energy Survey
• Finding strong lenses in CFHTLS using convolutional neural networks
• Fermi LAT Observations of Supernova Remnants Interacting with Molecular Clouds
• Typing Supernova Remnants Using X-ray Line Emission Morphologies
• Automatic unsupervised classification of all SDSS/DR7 galaxy spectra
• Identifying Strong Lenses with Unsupervised Machine Learning using Convolutional Autoencoder
• The low frequency radio emission and spectrum of the extended SNR W44: new VLA observations at 74 and 324 MHz
• The Mipsgal View of Supernova Remnants in the Galactic Plane
• AstroVaDEr: Astronomical Variational Deep Embedder for Unsupervised Morphological Classification of Galaxies and Synthetic Image Generation
• Unsupervised Classification of Variable Stars
• A Complete Catalogue of Dusty Supernova Remnants
• Automated detection of extended sources in radio maps: progress from the SCORPIO survey
• Interstellar gas towards the TeV gamma-ray sources HESS J1640-465 and HESS J1641-463
• Radio astronomical images object detection and segmentation: A benchmark on deep learning methods
• Evolutionary Map of the Universe (EMU):Compact radio sources in the SCORPIO field towards the Galactic plane
• X-ray Study of Spatial Structures in Tycho's Supernova Remnant Using Unsupervised Deep Learning