QUEST (Quasar Unsupervised Encoder and Synthesis Tool): A machine learning framework to generate quasar spectra
arXiv:2510.23206 · doi:10.1051/0004-6361/202557763
Abstract
Quasars at the redshift frontier (z > 7.0) are fundamental probes of black hole (BH) growth and evolution but notoriously difficult to identify. At these redshifts, machine learning-based selection methods have proven to be efficient, but require appropriate training sets to express their full potential. Here, we present QUEST, a Variational Auto-Encoder capable of generating realistic quasar spectra that can be post-processed for generating synthetic photometry and for spectral imputation. We start from the SDSS DR16Q catalogue, pre-process the spectra, and vet the sample to obtain a clean data set. After training the model, we investigate the properties of its latent space to understand whether it has learnt relevant physics. We provide a pipeline to generate photometry from the sampled spectra, compare it with actual quasar photometry, and showcase the capabilities of the model in reconstructing and extending quasar spectra. The trained network faithfully reproduces the input spectrum, both in terms of sample median and variance. By examining the latent space, we find correlations with continuum and bolometric luminosity, BH mass, redshift, continuum slope, and emission line properties. When used to generate photometry, we find results in excellent agreement with the control sample. The model provides satisfactory results in reconstructing emission lines: estimates of the BH mass from the reconstructed spectra are in good agreement with those from the original spectra. Furthermore, when spectra with broad absorption line features are reconstructed, the model successfully interpolates over the absorption systems. Compared with previous work, we find excellent agreement between the spectra sampled from our model and the output of their results. However, QUEST does not require any ad-hoc tuning, and is capable of reproducing the full variety of spectra available in the training set.
20 pages (15 main text + 1 references & aknowledgments + 4 appendix). Abstract abridged for Arxiv. Submitted to A&A, comments welcome! We release the tool as a python package, available at: https://github.com/cosmic-dawn-group/QUEST
References in corpus (61)
- Array Programming with NumPy
- Astropy: A Community Python Package for Astronomy
- The Astropy Project: Sustaining and Growing a Community-oriented Open-source Project and the Latest Major Release (v5.0) of the Core Package
- Coevolution (Or Not) of Supermassive Black Holes and Host Galaxies
- Binary companions of evolved stars in APOGEE DR14: Search method and catalog of ~5,000 companions
- The UKIRT Infrared Deep Sky Survey (UKIDSS)
- A luminous quasar at a redshift of z = 7.085
- Constraining the Evolution of the Ionizing Background and the Epoch of Reionization with z ~ 6 Quasars II: A Sample of 19 Quasars
- An 800-million-solar-mass black hole in a significantly neutral Universe at redshift 7.5
- The Assembly of the First Massive Black Holes
- Euclid. I. Overview of the Euclid mission
- The Sloan Digital Sky Survey Quasar Catalog: Sixteenth Data Release
- A Luminous Quasar at Redshift 7.642
- The Sloan Digital Sky Survey Quasar Catalog: twelfth data release
- Pōniuā'ena: A Luminous Quasar Hosting a 1.5 Billion Solar Mass Black Hole
- Hydrogen reionisation ends by : Lyman- optical depth measured by the XQR-30 sample
- An Analysis of the Shapes of Interstellar Extinction Curves. VII. Milky Way Spectrophotometric Optical-through-Ultraviolet Extinction and Its R-Dependence
- The origins of massive black holes
- The first ultraviolet quasar stacked spectrum at z=2.4 from WFC3
- Fe II Emission in 14 Low-Redshift Quasars: I - Observations
- The Black Hole Mass - Galaxy Bulge Relationship for QSOs in the SDSS DR3
- The Sloan Digital Sky Survey Reverberation Mapping Project: Sample Characterization
- A Catalog of Quasar Properties from Sloan Digital Sky Survey Data Release 16
- One Relation for All Wavelengths: The Far-Ultraviolet to Mid-Infrared Milky Way Spectroscopic R(V) Dependent Dust Extinction Relationship
- A Significantly Neutral Intergalactic Medium Around the Luminous z=7 Quasar J0252-0503
- An X-shooter composite of bright 1 < z < 2 quasars from UV to infrared
- FUSE Measurements of Far-Ultraviolet Extinction. III. The Dependence on R(V) and Discrete Feature Limits from 75 Galactic Sightlines
- The Kinematics of z ~ 6 Quasar Host Galaxies
- The X-shooter/ALMA Sample of Quasars in the Epoch of Reionization. II. Black Hole Masses, Eddington Ratios, and the Formation of the First Quasars
- The X-SHOOTER/ALMA sample of Quasars in the Epoch of Reionization. I. NIR spectral modeling, iron enrichment and broad emission line properties
- Milky Way Mid-Infrared Spitzer Spectroscopic Extinction Curves: Continuum and Silicate Features
- An unbiased measurement of the UV background and its evolution via the proximity effect in quasar spectra
- BAL and non-BAL quasars: continuum, emission and absorption properties establish a common parent sample
- A SPectroscopic survey of biased halos In the Reionization Era (ASPIRE): JWST Reveals a Filamentary Structure around a z=6.61 Quasar
- SpeX near-infrared spectroscopic extinction curves in the Milky Way
- SPECULATOR: Emulating stellar population synthesis for fast and accurate galaxy spectra and photometry
- Modelling type 1 quasar colours in the era of Rubin and Euclid
- Chronicling the reionization history at with emergent quasar damping wings
- A comparison of quasar emission reconstruction techniques for Lyman- and Lyman- transmission
- The Pan-STARRS1 z>5.6 quasar survey II: Discovery of 55 Quasars at 5.6<z<6.5
- The Pan-STARRS1 Quasar Survey: III. The Quasar Luminosity Function
- Constraining galaxy overdensities around three z~6.5 quasars with ALMA and MUSE
- The Physical Properties of Low Redshift FeLoBAL Quasars. I. Spectral Synthesis Analysis of the BAL Outflows using
- Random Forests as a viable method to select and discover high redshift quasars
- Autoencoding Galaxy Spectra I: Architecture
- Placing High-Redshift Quasars in Perspective: a Catalog of Spectroscopic Properties from the Gemini Near Infrared Spectrograph -- Distant Quasar Survey
- A robust estimator of mutual information for deep learning interpretability
- On the Connection between Supermassive Black Hole and Galaxy Growth in the Reionization Epoch
- A Mixture of LBG Overdensities in the Fields of Three Quasars: Implications for the Robustness of Photometric Selection
- Explaining dark matter halo density profiles with neural networks
- Euclid preparation. XXX. Performance assessment of the NISP Red-Grism through spectroscopic simulations for the Wide and Deep surveys
- Gemini Near Infrared Spectrograph -- Distant Quasar Survey: Augmented Spectroscopic Catalog and a Prescription for Correcting UV-Based Quasar Redshifts
- Molecular gas and dust properties in quasar hosts
- Paving the Way for Euclid and JWST via Optimal Selection of High-z Quasars
- The CatSouth Quasar Candidate Catalog for the Southern Sky and a Unified All-Sky Catalog Based on Gaia DR3
- A Generative Model for Quasar Spectra
- Discovery and characterization of 25 new quasars at 4.6 < z < 6.9 from wide-field multi-band surveys
- FNet II: Spectral Classification of Quasars, Galaxies, Stars, and broad absorption line (BAL) Quasars
- A complete search for redshift z>6.5 quasars in the VIKING survey
- Automated quasar continuum estimation using neural networks: a comparative study of deep-learning architectures
- Spectroscopic Quasar Anomaly Detection (SQuAD) I: Rest-Frame UV Spectra from SDSS DR16