English

Identifying Anomalous DESI Galaxy Spectra with a Variational Autoencoder

Instrumentation and Methods for Astrophysics 2026-03-12 v2 Cosmology and Nongalactic Astrophysics

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.

Keywords

Cite

@article{arxiv.2506.17376,
  title  = {Identifying Anomalous DESI Galaxy Spectra with a Variational Autoencoder},
  author = {C. Nicolaou and R. P. Nathan and O. Lahav and A. Palmese and A. Saintonge and J. Aguilar and S. Ahlen and C. Allende Prieto and S. Bailey and S. BenZvi and D. Bianchi and A. Brodzeller and D. Brooks and T. Claybaugh and A. de la Macorra and J. Della Costa and Arjun Dey and P. Doel and J. E. Forero-Romero and E. Gaztañaga and S. Gontcho A Gontcho and G. Gutierrez and K. Honscheid and C. Howlett and M. Ishak and R. Kehoe and D. Kirkby and T. Kisner and A. Kremin and A. Lambert and M. Landriau and L. Le Guillou and A. Meisner and R. Miquel and J. Moustakas and S. Nadathur and F. Prada and I. Pérez-Ràfols and G. Rossi and E. Sanchez and M. Schubnell and M. Siudek and D. Sprayberry and G. Tarlé and B. A. Weaver and H. Zou},
  journal= {arXiv preprint arXiv:2506.17376},
  year   = {2026}
}

Comments

20 pages, 18 figures, 1 table

R2 v1 2026-07-01T03:27:17.146Z