Applying Vision Transformers on Spectral Analysis of Astronomical Objects

Abstract

We apply pre-trained Vision Transformers (ViTs), originally developed for image recognition, to the analysis of astronomical spectral data. By converting traditional one-dimensional spectra into two-dimensional image representations, we enable ViTs to capture both local and global spectral features through spatial self-attention. We fine-tune a ViT pretrained on ImageNet using millions of spectra from the SDSS and LAMOST surveys, represented as spectral plots. Our model is evaluated on key tasks including stellar object classification and redshift ($z$) estimation, where it demonstrates strong performance and scalability. We achieve classification accuracy higher than Support Vector Machines and Random Forests, and attain $R^2$ values comparable to AstroCLIP's spectrum encoder, even when generalizing across diverse object types. These results demonstrate the effectiveness of using pretrained vision models for spectroscopic data analysis. To our knowledge, this is the first application of ViTs to large-scale, which also leverages real spectroscopic data and does not rely on synthetic inputs.

9 pages, 9 figures

References in corpus (31)

• The Gaia mission
• The Astropy Project: Sustaining and Growing a Community-oriented Open-source Project and the Latest Major Release (v5.0) of the Core Package
• The 2dF Galaxy Redshift Survey: Spectra and redshifts
• The Baryon Oscillation Spectroscopic Survey of SDSS-III
• The Apache Point Observatory Galactic Evolution Experiment (APOGEE)
• The First Data Release (DR1) of the LAMOST general survey
• The SDSS-IV extended Baryon Oscillation Spectroscopic Survey: Overview and Early Data
• Damped Lyman alpha Systems
• De Re Metallica: The cosmic chemical evolution of galaxies
• Galaxy Zoo 2: detailed morphological classifications for 304,122 galaxies from the Sloan Digital Sky Survey
• Asteroseismology of Solar-Type and Red-Giant Stars
• The Sloan Digital Sky Survey Quasar Catalog: twelfth data release
• The Cannon: A data-driven approach to stellar label determination
• DESI Bright Galaxy Survey: Final Target Selection, Design, and Validation
• J-PLUS: The Javalambre Photometric Local Universe Survey
• The VIMOS Public Extragalactic Redshift Survey (VIPERS). Full spectroscopic data and auxiliary information release (PDR-2)
• Gaia Data Release 3: Astrophysical parameters inference system (Apsis) I -- methods and content overview
• The Completed SDSS-IV extended Baryon Oscillation Spectroscopic Survey: Large-scale Structure Catalogs for Cosmological Analysis
• Calibrations of Atmospheric Parameters Obtained from the First Year of SDSS-III APOGEE Observations
• AstroCLIP: A Cross-Modal Foundation Model for Galaxies
• Machine learning technique for morphological classification of galaxies from the SDSS. I. Photometry-based approach
• ASTROMER: A transformer-based embedding for the representation of light curves
• The VIMOS Public Extragalactic Redshift Survey (VIPERS): spectral classification through Principal Component Analysis
• J-PLUS: Support Vector Machine Applied to STAR-GALAXY-QSOClassification
• Darth Fader: Using wavelets to obtain accurate redshifts of spectra at very low signal-to-noise
• Estimating Spectroscopic Redshifts by Using k Nearest Neighbors Regression I. Description of Method and Analysis
• Inferring stellar parameters and their uncertainties from high-resolution spectroscopy using invertible neural networks
• The LAMOST Complete Spectroscopic Survey of Pointing Area (LaCoSSPAr) in the Southern Galactic Cap I. The Spectroscopic Redshift Catalog
• Representation learning for automated spectroscopic redshift estimation
• SpecPT (Spectroscopy Pre-trained Transformer) Model for Extragalactic Spectroscopy: I. Architecture and Automated Redshift Measurement
• The Maunakea Spectroscopic Explorer: Thousands of Fibers, Infinite Possibilities