AI-Driven Reconstruction of Large-Scale Structure from Combined Photometric and Spectroscopic Surveys

2025/04/08 by W. J. Du, Xiaolin Luo, Du, Wenying +17
Physics and Astronomy · #Astronomy and Astrophysical Research #Cosmology and Nongalactic Astrophysics (astro-ph.CO) #FOS: Physical sciences #Galaxies: Formation, Evolution, Phenomena #Gamma-ray bursts and supernovae #Instrumentation and Methods for Astrophysics (astro-ph.IM)

paper · pdf · doi:10.48550/arxiv.2504.06309

openalex publication_date 2025/04/08 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Galaxy surveys are crucial for studying large-scale structure (LSS) and cosmology, yet they face limitations--imaging surveys provide extensive sky coverage but suffer from photo-z uncertainties, while spectroscopic surveys yield precise redshifts but are sample-limited. To take advantage of both photo-z and spec-z data while eliminating photo-z errors, we propose a deep learning framework based on a dual UNet architecture that integrates these two datasets at the field level to reconstruct the 3D photo-z density field. We train the network on mock samples representative of stage-IV spectroscopic surveys, utilizing CosmicGrowth simulations with a z=0.59 snapshot containing 2048³ particles in a (1200~h^-1\rm Mpc)³ volume. Several metrics, including correlation coefficient, MAE, MSE, PSNR, and SSIM, validate the model's accuracy. Moreover, the reconstructed power spectrum closely matches the ground truth at small scales (k \gtrsim 0.06~h/\rm Mpc) within the 1σ confidence level, while the UNet model significantly improves the estimation of photo-z power spectrum multipoles. This study demonstrates the potential of deep learning to enhance LSS reconstruction by using both spectroscopic and photometric data.

Citations

• Imputation of Missing Photometric Data and Photometric Redshift Estimation for CSST
• Measurement of the photometric Baryon Acoustic Oscillations with self-calibrated redshift distribution
• Volumetric Cloud Field Reconstruction
• CNN photometric redshifts in the SDSS at r≤ 20
• Hyper Suprime-Cam Year 3 Results: Cosmology from Cosmic Shear Two-point Correlation Functions
• Hyper Suprime-Cam Year 3 Results: Cosmology from Galaxy Clustering and Weak Lensing with HSC and SDSS using the Emulator Based Halo Model
• Higher-order statistics of the large-scale structure from photometric redshifts
• A deep learning method based on patchwise training for reconstructing temperature field
• Galaxy Correlation Function and Local Density from Photometric Redshifts Using the Stochastic Order Redshift Technique (SORT)
• GAlaxy Light profile convolutional neural NETworks (GaLNets). I. fast and accurate structural parameters for billion galaxy samples
• Neural networks and standard cosmography with newly calibrated high redshift GRB observations
• Dark Energy Survey Year 3 Results: A 2.7% measurement of Baryon Acoustic Oscillation distance scale at redshift 0.835
• Dark Energy Survey Year 3 Results: Cosmological Constraints from Galaxy Clustering and Weak Lensing
• Cosmic Velocity Field Reconstruction Using AI
• Photometric redshifts with machine learning, lights and shadows on a complex data science use case
• Dark Energy Survey Year 3 Results: Calibration of Lens Sample Redshift Distributions using Clustering Redshifts with BOSS/eBOSS
• Photometric Redshift Estimation with a Convolutional Neural Network: NetZ
• KiDS-1000 Cosmology: Multi-probe weak gravitational lensing and spectroscopic galaxy clustering constraints
• KiDS-1000 Cosmology: Cosmic shear constraints and comparison between two point statistics
• KiDS-1000 catalogue: Redshift distributions and their calibration
• Testing KiDS cross-correlation redshifts with simulations
• The PAU Survey: Photometric redshifts using transfer learning from simulations
• Cosmology with the Wide-Field Infrared Survey Telescope -- Multi-Probe Strategies
• The Clustering of DESI-like Luminous Red Galaxies Using Photometric Redshifts
• Representation learning for automated spectroscopic redshift estimation
• Cosmology from the Chinese Space Station Optical Survey (CSS-OS)
• Morpho-Photometric Redshifts
• Cosmology from cosmic shear power spectra with Subaru Hyper Suprime-Cam first-year data
• The LSST Dark Energy Science Collaboration (DESC) Science Requirements Document
• CosmicGrowth Simulations---Cosmological simulations for structure growth studies
• The many flavours of photometric redshifts
• Deep learning Approach for Classifying, Detecting and Predicting Photometric Redshifts of Quasars in the Sloan Digital Sky Survey Stripe 82
• Photometric redshifts for the Kilo-Degree Survey. Machine-learning analysis with artificial neural networks
• Stochastic Order Redshift Technique (SORT): a simple, efficient and robust method to improve cosmological redshift measurements
• Dark Energy Survey Year 1 Results: Cross-Correlation Redshifts - Methods and Systematics Characterization
• Dark Energy Survey Year 1 Results: Cosmological Constraints from Galaxy Clustering and Weak Lensing
• Dark Energy Survey Year 1 Results: Cosmological Constraints from Cosmic Shear
• Non-negative Matrix Factorization for Self-calibration of Photometric Redshift Scatter in Weak-lensing Surveys
• The-wiZZ: Clustering redshift estimation for everyone
• KiDS-450: Cosmological parameter constraints from tomographic weak gravitational lensing
• DNF - Galaxy photometric redshift by Directional Neighbourhood Fitting
• ANNz2 - photometric redshift and probability distribution function estimation using machine learning
• Wide-Field InfrarRed Survey Telescope-Astrophysics Focused Telescope Assets WFIRST-AFTA 2015 Report
• Sub-Megaparsec Individual Photometric Redshift Estimation from Cosmic Web Constraints
• Growth of Cosmic Structure: Probing Dark Energy Beyond Expansion
• Large-scale analysis of the SDSS-III DR8 photometric luminous galaxies angular correlation function
• Clustering-based redshift estimation: method and application to data
• CFHTLenS tomographic weak lensing cosmological parameter constraints: Mitigating the impact of intrinsic galaxy alignments
• Recovering Redshift Distributions with Cross-Correlations: Pushing The Boundaries
• On using angular cross-correlations to determine source redshift distributions
• NINE-YEAR WILKINSON MICROWAVE ANISOTROPY PROBE ( WMAP ) OBSERVATIONS: COSMOLOGICAL PARAMETER RESULTS
• Observational Probes of Cosmic Acceleration
• Acoustic scale from the angular power spectra of SDSS-III DR8 photometric luminous galaxies
• Clustering of Photometric Luminous Red Galaxies II: Cosmological Implications from the Baryon Acoustic Scale
• Galaxy Zoo Morphology and Photometric Redshifts in the Sloan Digital Sky Survey
• Resolving Subhaloes' Lives with the Hierarchical Bound-Tracing Algorithm
• Reconstructing Redshift Distributions with Cross-Correlations: Tests and an Optimized Recipe
• Photometric redshifts: estimating their contamination and distribution using clustering information
• Seven-Year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Cosmological Interpretation
• Photometric redshift estimation using Gaussian processes
• Self-calibration of photometric redshift scatter in weak-lensing surveys
• Power spectrum of the maxBCG sample: detection of acoustic oscillations using galaxy clusters
• New Approaches To Photometric Redshift Prediction Via Gaussian Process Regression In The Sloan Digital Sky Survey
• LSST: from Science Drivers to Reference Design and Anticipated Data Products
• Calibrating Redshift Distributions beyond Spectroscopic Limits with Cross‐Correlations
• The Correlation Function of Optically Selected Galaxy Clusters in the Sloan Digital Sky Survey
• Photometric Redshift Estimation on SDSS Data Using Random Forests
• Observational Tests of Modified Gravity
• Report of the Dark Energy Task Force
• Using Galaxy Two-point Correlation Functions to Determine the Redshift Distributions of Galaxies Binned by Photometric Redshift
• The Clustering of Luminous Red Galaxies in the Sloan Digital Sky Survey Imaging Data
• Accurate photometric redshifts for the CFHT legacy survey calibrated using the VIMOS VLT deep survey
• Novel Methods for Predicting Photometric Redshifts from Broad Band Photometry using Virtual Sensors
• A Census of Object Types and Redshift Estimates in the SDSS Photometric Catalog from a Trained Decision Tree Classifier
• Estimating Photometric Redshifts Using Support Vector Machines
• Evolutionary population synthesis: models, analysis of the ingredients and application to high-z galaxies
• Image quality assessment: from error visibility to structural similarity
• Photometric redshifts with the Multilayer Perceptron Neural Network: application to the HDF-S and SDSS
• Weak Gravitational Lensing by Large-Scale Structure
• Galaxy types in the Sloan Digital Sky Survey using supervised artificial neural networks
• Photometric Redshifts for the SDSS Early Data Release
• Estimating photometric redshifts with artificial neural networks
• Measuring the Redshift Evolution of Clustering: the Hubble Deep Field South
• Measuring and Modelling the Redshift Evolution of Clustering: the Hubble Deep Field North
• Bayesian photometric redshift estimation

Related

• Dark Energy Survey Year 3 results: Simulation-based wCDM inference from weak lensing and galaxy clustering maps with deep learning: Analysis design
• Hybrid-z: Enhancing Kilo-Degree Survey bright galaxy sample photometric redshifts with deep learning
• Deep learning based photometric redshifts for the Kilo-Degree Survey Bright Galaxy Sample
• From Redshift to Real Space: Combining Linear Theory With Neural Networks
• Deep Learning Improves Photometric Redshifts in All Regions of Color Space