The PAU Survey: Photometric redshifts using transfer learning from simulations

Abstract

In this paper we introduce the \textsc{Deepz} deep learning photometric redshift (photo-$z$) code. As a test case, we apply the code to the PAU survey (PAUS) data in the COSMOS field. \textsc{Deepz} reduces the ${\sigma }_{68}$ scatter statistic by 50\% at $i_{\mathrm{AB}}=22.5$ compared to existing algorithms. This improvement is achieved through various methods, including transfer learning from simulations where the training set consists of simulations as well as observations, which reduces the need for training data. The redshift probability distribution is estimated with a mixture density network (MDN), which produces accurate redshift distributions. Our code includes an autoencoder to reduce noise and extract features from the galaxy SEDs. It also benefits from combining multiple networks, which lowers the photo-$z$ scatter by 10 percent. Furthermore, training with randomly constructed coadded fluxes adds information about individual exposures, reducing the impact of photometric outliers. In addition to opening up the route for higher redshift precision with narrow bands, these machine learning techniques can also be valuable for broad-band surveys.

arXiv commentAccepted version

References in corpus (53)

• Planck 2015 results. XIII. Cosmological parameters
• Galactic Stellar and Substellar Initial Mass Function
• Gaia Data Release 2. Summary of the contents and survey properties
• Star Formation in Galaxies Along the Hubble Sequence
• The Dust Content and Opacity of Actively Star-Forming Galaxies
• A systematic study of the class imbalance problem in convolutional neural networks
• The propagation of uncertainties in stellar population synthesis modeling I: The relevance of uncertain aspects of stellar evolution and the IMF to the derived physical properties of galaxies
• EAZY: A Fast, Public Photometric Redshift Code
• Accurate photometric redshifts for the CFHT Legacy Survey calibrated using the VIMOS VLT Deep Survey
• The u'g'r'i'z' Standard Star Network
• Astrophysics Source Code Library
• NYU-VAGC: a galaxy catalog based on new public surveys
• COSMOS Photometric Redshifts with 30-bands for 2-deg2
• Bayesian photometric redshift estimation
• The propagation of uncertainties in stellar population synthesis modeling III: model calibration, comparison, and evaluation
• The COSMOS2015 Catalog: Exploring the 1<z<6 Universe with half a million galaxies
• The Galaxy Luminosity Function and Luminosity Density at Redshift z=0.1
• zCOSMOS: A Large VLT/VIMOS redshift survey covering 0 < z < 3 in the COSMOS field
• LSST Science Book, Version 2.0
• ANNz: estimating photometric redshifts using artificial neural networks
• The 2013 Release of Cloudy
• Weak Gravitational Lensing and its Cosmological Applications
• Gravitational Lensing Analysis of the Kilo Degree Survey
• Nebular Continuum and Line Emission in Stellar Population Synthesis Models
• The VIMOS VLT Deep Survey - First epoch VVDS-Deep survey: 11564 spectra with 17.5<=IAB<=24, and the redshift distribution over 0< z <=5
• The fourth data release of the Kilo-Degree Survey: ugri imaging and nine-band optical-IR photometry over 1000 square degrees
• Calibrating Redshift Distributions Beyond Spectroscopic Limits with Cross-Correlations
• PHAT: PHoto-z Accuracy Testing
• Mapping the Galaxy Color-Redshift Relation: Optimal Photometric Redshift Calibration Strategies for Cosmology Surveys
• Dark Energy Survey Year 1 Results: Redshift distributions of the weak lensing source galaxies
• Ameliorating Systematic Uncertainties in the Angular Clustering of Galaxies: A Study using SDSS-III
• Generating Images with Perceptual Similarity Metrics based on Deep Networks
• ArborZ: Photometric Redshifts Using Boosted Decision Trees
• Photometric redshifts with the Multilayer Perceptron Neural Network: application to the HDF-S and SDSS
• The Complete Calibration of the Color-Redshift Relation (C3R2) Survey: Analysis and Data Release 2
• DNF - Galaxy photometric redshift by Directional Neighbourhood Fitting
• Modelling the number density of Halpha emitters for future spectroscopic near-IR space missions
• Using Galaxy Two-point Correlation Functions to Determine the Redshift Distributions of Galaxies Binned by Photometric Redshift
• Cross-Correlation of spectroscopic and photometric galaxy surveys: cosmology from lensing and redshift distortions
• Photometric redshifts for the Kilo-Degree Survey. Machine-learning analysis with artificial neural networks
• Using neural networks to estimate redshift distributions. An application to CFHTLenS
• Precise photometric redshifts with a narrow-band filter set: The PAU Survey at the William Herschel Telescope
• The PAU Survey: Early demonstration of photometric redshift performance in the COSMOS field
• Data augmentation for machine learning redshifts applied to SDSS galaxies
• Photo-z Performance for Precision Cosmology
• Analysis of a Custom Support Vector Machine for Photometric Redshift Estimation and the Inclusion of Galaxy Shape Information
• The Physics of the Accelerating Universe Camera
• The PAU Survey: star-galaxy classification with multi narrow-band data
• Parametrising Star Formation Histories
• Combining spectroscopic and photometric surveys using angular cross-correlations II: Parameter constraints from different physical effects
• The PAU Survey: Operation and orchestration of multi-band survey data
• Photo-z Quality Cuts and their Effect on the Measured Galaxy Clustering
• Data Mining in Clinical Trial Text: Transformers for Classification and Question Answering Tasks

Cited by in corpus (24)

• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• Photometric Redshift Estimation with a Convolutional Neural Network: NetZ
• The PAU Survey: An improved photo-$z$ sample in the COSMOS field
• Improving the reliability of photometric redshift with machine learning
• The evolution of barred galaxies in the EAGLE simulations
• Extracting cosmological parameters from N-body simulations using machine learning techniques
• The PAU Survey: Measurement of Narrow-band galaxy properties with Approximate Bayesian Computation
• The PAU Survey: narrowband photometric redshifts using Gaussian processes
• The PAU Survey: Intrinsic alignments and clustering of narrow-band photometric galaxies
• The PAU survey: Estimating galaxy photometry with deep learning
• TOPz: Photometric redshifts for J-PAS
• The PAU Survey: Measurements of the 4000 Å spectral break with narrow-band photometry
• Transient Classifiers for Fink: Benchmarks for LSST
• PICZL: Image-based Photometric Redshifts for AGN
• The VVV near-IR galaxy catalogue in a Northern part of the Galactic disc
• The PAU Survey: Photometric redshift estimation in deep wide fields
• The PAU Survey: Narrow-band image photometry
• Euclid: Photometric redshift calibration with the clustering redshifts technique
• The PAU Survey: The quasar UV and Ly$\alpha$ luminosity functions at $2.7<z<5.3$
• Euclid: Photometric redshift calibration with self-organising maps
• Machine learning applications in astrophysics: Photometric redshift estimation
• The PAU Survey: Classifying low-z SEDs using Machine Learning clustering
• Non-Sequential Neural Network for Simultaneous, Consistent Classification and Photometric Redshifts of OTELO Galaxies
• Power Spectrum Emulators from Neural Networks and Tree-Based Methods