DeepRed: an architecture for redshift estimation
Alessandro Meroni, Nicolò Oreste Pinciroli Vago, Piero Fraternali
TL;DR
Redshift estimation from astronomical images is costly and constrained by dataset heterogeneity. DeepRed introduces a pipeline that ensembles generic computer-vision backbones—such as EfficientNet, Swin Transformer, and MLP-Mixer—via a linear regression ensemble on latent outputs to robustly predict redshift for galaxies, gravitational lenses, and gravitationally-lensed transients, while incorporating SHAP-based explainability. Across four simulated DeepGraviLens datasets and real KiDS and SDSS data, DeepRed achieves state-of-the-art NMAD and $\sigma_{68}$ reductions (up to $55\%$ and $50\%$, respectively) and lower bias/outlier rates, with SHAP localization exceeding $95\%$ accuracy on high-quality images. The results demonstrate strong generalization across morphologies and observational conditions, supporting scalable redshift estimation for upcoming sky surveys; future work includes cross-instrument validation and integration of time-domain data.
Abstract
Estimating redshift is a central task in astrophysics, but its measurement is costly and time-consuming. In addition, current image-based methods are often validated on homogeneous datasets. The development and comparison of networks able generalize across different morphologies, ranging from galaxies to gravitationally-lensed transients, and observational conditions, remain an open challenge. This work proposes DeepRed, a deep learning pipeline that demonstrates how modern computer vision architectures, including ResNet, EfficientNet, Swin Transformer, and MLP-Mixer, can estimate redshifts from images of galaxies, gravitational lenses, and gravitationally-lensed supernovae. We compare these architectures and their ensemble to both neural networks (A1, A3, NetZ, and PhotoZ) and a feature-based method (HOG+SVR) on simulated (DeepGraviLens) and real (KiDS, SDSS) datasets. Our approach achieves state-of-the-art results on all datasets. On DeepGraviLens, DeepRed achieves a significant improvement in the Normalized Mean Absolute Deviation compared to the best baseline (PhotoZ): 55% on DES-deep (using EfficientNet), 51% on DES-wide (Ensemble), 52% on DESI-DOT (Ensemble), and 46% on LSST-wide (Ensemble). On real observations from the KiDS survey, the pipeline outperforms the best baseline (NetZ), improving NMAD by 16% on a general test set without high-probability lenses (Ensemble) and 27% on high-probability lenses (Ensemble). For non-lensed galaxies in the SDSS dataset, the MLP-Mixer architecture achieves a 5% improvement over the best baselines (A3 and NetZ). SHAP shows that the models correctly focus on the objects of interest with over 95% localization accuracy on high-quality images, validating the reliability of the predictions. These findings suggest that deep learning is a scalable, robust, and interpretable solution for redshift estimation in large-scale surveys.