astro-ph.IMOct 28, 2026

UnEZ: A Fully Unsupervised Self-Templating Neural Network for Redshift and Line Strength Estimation

Authors: Asha Abd, Karl Glazebrook, Colin Jacobs

Organizations: Centre for Astrophysics and Supercomputing, Swinburne University of Technology, PO Box 218, Hawthorn, VIC 3122, Australia

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

As spectroscopic surveys grow, estimating galaxy properties such as metallicity, star formation rate, and redshift becomes computationally cumbersome. Methods of estimating such properties rely on template spectra or curated redshift catalogs, which fail to represent the diversity of objects seen in large-scale surveys. We present UnEZ, a fully unsupervised, physics-informed machine learning model that can predict spectroscopic redshifts and line strengths with no prior redshift or line strength labels. An encoder predicts emission and absorption line amplitudes/widths, and a fixed deterministic decoder constructs a unique template, estimated redshift, redshift confidence curve, and spectral model. UnEZ was tested on the Sloan Digital Sky Survey (SDSS) Data Release 18 due to its large dataset and thoroughly vetted redshift catalog which serves as an independent benchmark. Despite the lack of templates, line strength labels, or redshift labels, UnEZ's performance competes with modern redshift estimation algorithms. At SNR > 10, UnEZ achieves an outlier fraction of 0.459% and a normalized median absolute deviation (NMAD) of 1.237×10−41.237 \times 10^{-4}. At SNR > 3, UnEZ achieves an outlier fraction of 3.89% and an NMAD of 1.471×10−41.471 \times 10^{-4}. Furthermore, we demonstrate how UnEZ's line strength and redshift predictions can be utilized for various purposes such as constructing template spectra required in traditional redshift estimation algorithms. UnEZ illustrates how a fully unsupervised neural network can allow redshift, among other galaxy properties, to emerge naturally from the spectra themselves, providing a template-free, label-free approach that scales to the demands of future surveys.

Figures & tables

Figure 1: Diagram of the UnEZ neural network. The blue embedding block has learnable parameters while orange blocks are fixed deterministic functions with no learnable parameters.
Figure 2: Diagram of the embedding layer in Figure 1 . The red blocks contain a max-pooling layer while the cyan blocks do not. The last cyan blocks feed into the multi-headed attention mechanism, with the right-side block being the keys and the left-side block being the queries. Detailed specifications and hyperparameters are listed in Table 1 .
StageLayerConfigurationOutput shapeParameters
InputInputLayer—(N, 9400, 1)(N,\ 9400,\ 1)0
Conv Block 1Conv1D64 filters, k=5k=5(N, 9400, 64)(N,\ 9400,\ 64)384384
MaxPooling1Dpool size =2=2(N, 4700, 64)(N,\ 4700,\ 64)0
Dropoutrate =0.2=0.2(N, 4700, 64)(N,\ 4700,\ 64)0
Conv Block 2Conv1D128 filters, k=11k=11(N, 4700, 128)(N,\ 4700,\ 128)90 24090\,240
MaxPooling1Dpool size =2=2(N, 2350, 128)(N,\ 2350,\ 128)0
Table 1: Architecture of the UnEZ model ( NN = batch size). All Conv1D layers use same padding and ReLU activation. A uniform dropout rate p=0.2p=0.2 is applied after every Conv1D. The line branch provides the attention query; The depth branch provides the key and value.
 SDSSBOSS
Native Min/Max Wavelength3800Å–9200Å3600Å–10400Å
Native Resolution1500 at 3800Å, 2500 at 9000Å
Native Pixel Spacing10−410^{-4} dex
Resampled Pixel Spacing4.84 ×10−5\times 10^{-5} dex
Resampled Min/Max Wavelength3600Å– 9400Å
Table 2: Native and resampled spectral properties including resolution and wavelength range, and pixel spacing.
Figure 3: (top) Raw SDSS spectrum. (middle) Spectrum with bad pixels such as skylines, zero variance pixels, and cosmic rays removed. (bottom) Final preprocessed spectrum with the continuum removed that has been interpolated into uniform wavelength bins and clipped at the ends to remove edge effects. Bad pixels are denoted by the gray shaded region.
Figure 4: Comparison of SDSS catalog redshifts with UnEZ predictions for objects with 0.003<z<1.00.003<z<1.0 and SNR >10>10 . Top: Distribution of catalog redshifts, highlighting the dominance of galaxies (blue) and the increasing prevalence of quasars (orange) at higher redshift. Middle: Predicted versus catalog redshift, with the red dashed line marking the 1:1 relation. The concentration along this line indicates accurate recovery, while horizontal and parallel features mark common misidentifications. Bottom: Δζ\Delta\zeta and Δv\Delta\text{v} (in [ km⋅s−1\text{km}\cdot\text{s}^{-1} ]) versus catalog redshift showing how the spread trends with increasing redshift. UnEZ achieves NMAD = 1.24×10−41.24\times 10^{-4} with an outlier fraction of 0.46% across 771,086 spectra.
Figure 5: As Figure 4 , but for the lower SNR >3>3 sample. All panels, colors, and symbols are identical to Figure 4 . At this cut, UnEZ achieves NMAD = 1.47×10−41.47\times 10^{-4} with an outlier fraction of 3.89% across 1,817,847 spectra. Skyline contamination becomes markedly more prevalent as noise increases, and the strongest line-confusion degeneracies appear as ridges parallel to the 1:1 relation.
Figure 6: (top) Outlier fraction as a function of signal-to-noise ratio. (middle) NMAD as a function of SNR. The dashed lines in these plots correspond to NMAD and outlier fraction values shown in the main plot of Figure 4 . (bottom) Distribution of SNR across the entire dataset, with the dashed line representing the cutoff for how many points are needed for a meaningful measure of outlier fraction and NMAD and their errors. Performance quickly degrades with lower SNR regardless of quasar (orange) or galaxy (blue) classification.
Figure 7: Distribution of Δζ\Delta\zeta for galaxies (blue) and quasars (orange). The outlier threshold is highlighted by the red dashed line at Δζ=0.0015\Delta\zeta=0.0015 or Δ\Delta v = 450 km⋅s−1\text{km}\cdot\text{s}^{-1} .
ClassSample SizeInlier %NMAD
QSO45,51697.1%4.112×10−4 (123 km⋅s−1)4.112\times 10^{-4}\ (123\ \text{km}\cdot\text{s}^{-1})
Galaxy725,39199.7%1.194×10−4 (36 km⋅s−1)1.194\times 10^{-4}\ (36\ \text{km}\cdot\text{s}^{-1})
Total771,08699.5%1.237×10−4 (37 km⋅s−1)1.237\times 10^{-4}\ (37\ \text{km}\cdot\text{s}^{-1})
Table 3: Sample size, inlier fraction, and NMAD of each spectral class at SNR > 10.
ClassSample SizeInlier %NMAD
QSO147,32694.5%3.471×10−4 (104 km⋅s−1)3.471\times 10^{-4}\ (104\ \text{km}\cdot\text{s}^{-1})
Galaxy1,670,06996.3%1.407×10−4 (42 km⋅s−1)1.407\times 10^{-4}\ (42\ \text{km}\cdot\text{s}^{-1})
Total1,817,84796.1%1.471×10−4 (44 km⋅s−1)1.471\times 10^{-4}\ (44\ \text{km}\cdot\text{s}^{-1})
Table 4: As Table 3 , but for the SNR > 3 sample.
Figure 8: Inlier examples. (top) Starburst Galaxy. (middle) Quasar. (bottom) Quiescent Galaxy. SDSS spectra are shown in black while UnEZ’s spliced template is shown in red. The right panels show the SDSS photometry for each object with the fiber placement overlaid in cyan.
Figure 9: As Figure 8 , but for outliers. (top) Superimposed objects at different redshifts with multiple distinct sources in the photometry. (middle) Iron low-ionization broad absorption line (FeLoBAL) quasar spectrum. (bottom) Superimposed objects at different redshifts with the second source not visible in the SDSS photometry.
Figure 10: UnEZ’s predicted loss landscape for superimposed galaxies shown in the bottom panel of Figure 9 .
CategoryNNPercentage
SDSS correct62563.8%
UnEZ correct12512.8%
Both (superimposed)889.0%
Close585.9%
Neither515.2%
Unknown333.4%
Table 5: Visual classification of 980 inspected UnEZ redshift outliers.
Figure 11: Comparison of emission line fluxes predicted by UnEZ ( FUnEZF_{\rm UnEZ} ) with those from the SDSS catalog ( FSDSSF_{\rm SDSS} ) for H β\beta , [O III] λ\lambda 5008, H α\alpha , and the [N II] λλ\lambda\lambda 6550,6585 doublet. Only redshift inliers with SDSS line SNR > 5 and a non-zero UnEZ amplitude are shown. NN gives the number of objects in each panel. UnEZ fluxes are computed using Equation 14 . The [N II] doublet is the sum of both line fluxes in each catalog. The color bar shows the normalized count per bin. The thick and thin contours enclose 68% and 95% of the objects, while the dashed line marks the 1:1 relation.
Figure 12: BPT Diagram composed from UnEZ’s predicted line strength ratios with SDSS catalog line strength ratios overlaid. The red dashed line is an AGN diagnostic line separating AGN from starburst galaxies ( Kewley et al., 2001 ) .
Figure 13: SOM map showing redshift topology across the map indices. The color represents the median zUnEZz_{\text{UnEZ}} of the spectra belonging to that node. The red scatter points, labeled by their corresponding template spectrum, are the cluster centroids of the map’s node weights, while the white outlines show the cluster boundaries.
Figure 14: SOM map showing velocity dispersion topology across the map indices. The color represents the median velocity dispersion, predicted by UnEZ, of the spectra belonging to that node. The red scatter points, labeled by their corresponding template spectrum, are the cluster centroids, while the white outlines show the cluster boundaries.
Figure 15: Three templates constructed from median spectrum of each cluster predicted by UnEZ line strengths.
QSO →\rightarrow QSOGalaxy →\rightarrow GalaxyQSO →\rightarrow GalaxyGalaxy →\rightarrow QSO
13,778475,5339291083
Table 6: Table of SDSS classifications compared to UnEZ classifications.
Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

#Speciesλvac\lambda_{\rm vac} (Å)Cl.#Speciesλvac\lambda_{\rm vac} (Å)Cl.
1O vi1033.82B23H γ\gamma4341.68B
2Ly α\alpha1215.24B24[O iii ]4364.44E
3N v1240.81B25H β\beta4862.68B
4O i1305.53E26[O iii ]4932.60E
5C ii1335.31E27[O iii ]4960.30E
6Si iv1397.61E28[O iii ]5008.24E
Appendix
Table 7: Spectral lines used as physics priors in UnEZ. Wavelengths are rest-frame vacuum values from the SDSS spectro1d line table ( Sloan Digital Sky Survey, 2008 ) . Classifications: E = emission-only, A = absorption-only, B = emission or absorption.
Figure 16: As Figure 5 but for the brute force method. Line confusion degeneracies and skyline contamination become much more prevalent using the brute force method.
Figure 17: As Figure 6 with the brute force method plotted as well. Only 100,000 random spectra were selected for each method due to the long compute time required for the brute force method. UnEZ consistently performs better across the entire SNR range and converges toward peak performance at a lower SNR.
More in this topic

Astrophysics

See all 33 papers

Explore similar work

CardsList
Jun 11, 2026

Multi-Variable Stellar Parameter Estimation Using Residual Multitask Neural Networks

We present an end-to-end pipeline for estimating stellar parameters from Sloan Digital Sky Survey Data Release 12 spectra using a fully connected multitask neural network with residual blocks, whose hyperparameters are tuned via Bayesian optimization. The preprocessing pipeline includes per-spectrum standardization, RobustScaler normalization of the target variables -- effective temperature TeffT_{\mathrm{eff}}, metallicity [Fe/H][\mathrm{Fe/H}], and surface gravity log⁡g\log g -- and data augmentation via Gaussian noise injection. On a held-out test set, the model achieved Mean Absolute Errors (MAE) of 59.76 K59.76~\mathrm{K} for TeffT_{\mathrm{eff}}, 0.103 dex0.103~\mathrm{dex} for [Fe/H][\mathrm{Fe/H}], and 0.130 dex0.130~\mathrm{dex} for log⁡g\log g. Normalized against the full-scale range of each parameter, these results represent range-normalized errors between 1%1\% and 3%3\%, achieved with a highly efficient model complexity of approximately 540,000 trainable parameters. These results demonstrate that a compact residual multitask architecture, combined with principled signal preprocessing, provides a parameter-efficient solution for nonlinear parameter estimation in large-scale spectral datasets. In particular, the proposed model achieves competitive performance with substantially lower complexity than deeper neural network baselines.

Bruno Santos Meneses Barreto, Marcio Eisencraft

Escola Polit´ecnica, Universidade de S˜ao Paulo, SP

Parameter EstimationMulti-Task Learning
AskPDF
Jun 8, 2026

Integral Field Unit Spectroscopy with One Fiber

Integral field unit (IFU) spectroscopy provides spatially resolved spectra across galaxies, offering crucial insights into their evolution. However, its high observational cost limits current IFU datasets to ∼104\sim 10^4 objects. We present a multi-modal, probabilistic foundation model that predicts high-resolution spectra with calibrated uncertainties at arbitrary spatial locations within a galaxy directly from broadband images. Built on a masked autoencoder framework, our architecture injects fiber positional encodings and redshift aware wavelength encodings, enabling spatially conditioned predictions. Trained on 4.7 million images and single fiber spectroscopic observations from the Dark Energy Spectroscopic Instrument (DESI) survey, our model exploits the natural variance of fiber placements and the morphological self-similarity of galaxies to achieve IFU-like capabilities without any IFU training data. Predicted emission line flux maps match independent IFU observations from the Mapping Nearby Galaxies at APO (MaNGA) survey, with performance comparable to a supervised baseline trained directly on IFU data.

Zehao Peng, Biprateep Dey, Chris J. Maddison +1

University of Toronto · 2Vector Institute

Topic pending
AskPDF
Mar 25, 2026

SM-Net: Learning a Continuous Spectral Manifold from Multiple Stellar Libraries

We present SM-Net, a machine-learning model that learns a continuous spectral manifold from multiple high-resolution stellar libraries. SM-Net generates stellar spectra directly from the fundamental stellar parameters effective temperature (Teff), surface gravity (log g), and metallicity (log Z). It is trained on a combined grid derived from the PHOENIX-Husser, C3K-Conroy, OB-PoWR, and TMAP-Werner libraries. By combining their parameter spaces, we construct a composite dataset that spans a broader and more continuous region of stellar parameter space than any individual library. The unified grid covers Teff = 2,000-190,000 K, log g = -1 to 9, and log Z = -4 to 1, with spectra spanning 3,000-100,000 Angstrom. Within this domain, SM-Net provides smooth interpolation across heterogeneous library boundaries. Outside the sampled region, it can produce numerically smooth exploratory predictions, although these extrapolations are not directly validated against reference models. Zero or masked flux values are treated as unknowns rather than physical zeros, allowing the network to infer missing regions using correlations learned from neighbouring grid points. Across 3,538 training and 11,530 test spectra, SM-Net achieves mean squared errors of 1.47 x 10^-5 on the training set and 2.34 x 10^-5 on the test set in the transformed log1p-scaled flux representation. Inference throughput exceeds 14,000 spectra per second on a single GPU. We also release the model together with an interactive web dashboard for real-time spectral generation and visualisation. SM-Net provides a fast, robust, and flexible data-driven complement to traditional stellar population synthesis libraries.

Omar Anwar, Aaron S. G. Robotham, Luca Cortese +1

International Centre for Radio Astronomy Research (ICRAR) The University of Western Australia (UWA), 35 Stirling Hwy, Crawley WA 6009, Australia

AstrophysicsNeural Surrogate ModelingConditional Generative Modeling
AskPDF