Identifying Anomalous DESI Galaxy Spectra with a Variational Autoencoder

Abstract

The tens of millions of spectra being captured by the Dark Energy Spectroscopic Instrument (DESI) provide tremendous discovery potential. In this work we show how Machine Learning, in particular Variational Autoencoders (VAE), can detect anomalies in a sample of approximately 200,000 DESI spectra comprising galaxies, quasars and stars. We demonstrate that the VAE can compress the dimensionality of a spectrum by a factor of 100, while still retaining enough information to accurately reconstruct spectral features. We then detect anomalous spectra as those with high reconstruction error and those which are isolated in the VAE latent representation. The anomalies identified fall into two categories: spectra with artefacts and spectra with unique physical features. Awareness of the former can help to improve the DESI spectroscopic pipeline; whilst the latter can lead to the identification of new and unusual objects. To further curate the list of outliers, we use the Astronomaly package which employs Active Learning to provide personalised outlier recommendations for visual inspection. In this work we also explore the VAE latent space, finding that different object classes and subclasses are separated despite being unlabelled. We demonstrate the interpretability of this latent space by identifying tracks within it that correspond to various spectral characteristics. For example, we find tracks that correspond to increasing star formation and increase in broad emission lines along the Balmer series. In upcoming work we hope to apply the methods presented here to search for both systematics and astrophysically interesting objects in much larger datasets of DESI spectra.

20 pages, 18 figures, 1 table

References in corpus (33)

• Overview of the DESI Legacy Imaging Surveys
• A Unifying Review of Deep and Shallow Anomaly Detection
• Overview of the Instrumentation for the Dark Energy Spectroscopic Instrument
• The Early Data Release of the Dark Energy Spectroscopic Instrument
• The 2dF Galaxy Redshift Survey: Spectral Types and Luminosity Functions
• Spectral Classification of Quasars in the Sloan Digital Sky Survey: Eigenspectra; Redshift and Luminosity Effects
• The Spectroscopic Data Processing Pipeline for the Dark Energy Spectroscopic Instrument
• The 2dF Galaxy Redshift Survey: galaxy clustering per spectral type
• Distributions of Galaxy Spectral Types in the Sloan Digital Sky Survey
• The Robotic Multi-Object Focal Plane System of the Dark Energy Spectroscopic Instrument (DESI)
• The Target-selection Pipeline for the Dark Energy Spectroscopic Instrument
• The weirdest SDSS galaxies: results from an outlier detection algorithm
• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• The DESI Survey Validation: Results from Visual Inspection of the Quasar Survey Spectra
• Dimensionality Reduction of SDSS Spectra with Variational Autoencoders
• Astronomaly: Personalised Active Anomaly Detection in Astronomical Data
• Principal Component Analysis Of Synthetic Galaxy Spectra
• Galaxy spectral parameterisation in the 2dF Galaxy Redshift Survey as a diagnostic of star formation history
• Performance of the Quasar Spectral Templates for the Dark Energy Spectroscopic Instrument
• Unusual quasars from the Sloan Digital Sky Survey selected by means of Kohonen self-organising maps
• Objective Classification of Galaxy Spectra using the Information Bottleneck Method
• A Deep Learning Approach for Active Anomaly Detection of Extragalactic Transients
• Qualitative interpretation of galaxy spectra
• Characterising the target selection pipeline for the Dark Energy Spectroscopic Instrument Bright Galaxy Survey
• Exploring the Spectral Space of Low Redshift QSOs
• Decoding the spectra of SDSS early-type galaxies: New indicators of age and recent star formation
• Neural network-based anomaly detection for high-resolution X-ray spectroscopy
• The Optical Corrector for the Dark Energy Spectroscopic Instrument
• De-noising of galaxy optical spectra with autoencoders
• Investigation of stellar magnetic activity using variational autoencoder based on low-resolution spectroscopic survey
• A Physics-Informed Variational Autoencoder for Rapid Galaxy Inference and Anomaly Detection
• Identifying Quasars from the DESI Bright Galaxy Survey
• Finding Pegasus: Enhancing Unsupervised Anomaly Detection in High-Dimensional Data using a Manifold-Based Approach