SpecPT (Spectroscopy Pre-trained Transformer) Model for Extragalactic Spectroscopy: I. Architecture and Automated Redshift Measurement
arXiv:2501.01070 · doi:10.3847/1538-4357/ade053
Abstract
We introduce the Spectroscopy Pre-trained Transformer (SpecPT), a transformer-based model designed to analyze spectroscopic data, with applications in spectrum reconstruction and redshift measurement. Using the Early Data Release (EDR) of the DESI survey, we evaluate SpecPT's performance on two distinct datasets: the Bright Galaxy Survey (BGS) and Emission Line Galaxy (ELG) samples. SpecPT successfully reconstructs spectra, accurately capturing emission lines, absorption features, and continuum shapes while effectively reducing noise. For redshift prediction, SpecPT achieves competitive accuracy, with Normalized Median Absolute Deviation (NMAD) values of 0.0006 and 0.0008, and catastrophic outlier fractions of 0.20% and 0.80% for BGS and ELG, respectively. Notably, SpecPT performs consistently well across the full redshift range (), demonstrating its versatility and robustness. By leveraging its learned latent representations, SpecPT lays the groundwork for a foundational spectroscopic model, with potential applications in outlier detection, interstellar medium (ISM) property estimation, and transfer learning to other datasets. This work represents a first step in building a generalized framework for spectroscopic analysis, capable of scaling to the full DESI dataset and beyond.
17 pages, 13 figures. Submitted to ApJ
References in corpus (21)
- The Galaxy Evolution Explorer: A Space Ultraviolet Survey Mission
- The 2dF Galaxy Redshift Survey: Spectra and redshifts
- COSMOS Photometric Redshifts with 30-bands for 2-deg2
- Spectral Classification and Redshift Measurement for the SDSS-III Baryon Oscillation Spectroscopic Survey
- Understanding Galaxy Evolution through Emission Lines
- ANNz: estimating photometric redshifts using artificial neural networks
- GEMS: Galaxy Evolution from Morphologies and SEDs
- TPZ : Photometric redshift PDFs and ancillary information by using prediction trees and random forests
- DESI Bright Galaxy Survey: Final Target Selection, Design, and Validation
- The Spectroscopic Data Processing Pipeline for the Dark Energy Spectroscopic Instrument
- Target Selection and Validation of DESI Emission Line Galaxies
- Galaxy And Mass Assembly (GAMA): autoz spectral redshift measurements, confidence and errors
- An Application of Deep Neural Networks in the Analysis of Stellar Spectra
- ArborZ: Photometric Redshifts Using Boosted Decision Trees
- Estimating Photometric Redshifts Using Support Vector Machines
- Photometric Redshift Estimation with a Convolutional Neural Network: NetZ
- AstroCLIP: A Cross-Modal Foundation Model for Galaxies
- GAz: A Genetic Algorithm for Photometric Redshift Estimation
- Autoencoding Galaxy Spectra I: Architecture
- Autoencoding Galaxy Spectra II: Redshift Invariance and Outlier Detection
- Galaxy Spectra neural Network (GaSNet). II. Using Deep Learning for Spectral Classification and Redshift Predictions