Improving Photometric Redshift Estimates with Training Sample Augmentation

Moskowitz, Irene; Gawiser, Eric; Crenshaw, John Franklin; Andrews, Brett H.; Malz, Alex I.; Schmidt, Samuel

doi:10.3847/2041-8213/ad4039

Improving Photometric Redshift Estimates with Training Sample Augmentation

Journal Article · Mon May 13 00:00:00 EDT 2024 · The Astrophysical Journal. Letters

DOI:https://doi.org/10.3847/2041-8213/ad4039· OSTI ID:2349559

; ; ; ; ;

Abstract

Large imaging surveys will rely on photometric redshifts (photo- z 's), which are typically estimated through machine-learning methods. Currently planned spectroscopic surveys will not be deep enough to produce a representative training sample for Legacy Survey of Space and Time (LSST), so we seek methods to improve the photo- z estimates that arise from nonrepresentative training samples. Spectroscopic training samples for photo- z 's are biased toward redder, brighter galaxies, which also tend to be at lower redshift than the typical galaxy observed by LSST, leading to poor photo- z estimates with outlier fractions nearly 4 times larger than for a representative training sample. In this Letter, we apply the concept of training sample augmentation, where we augment simulated nonrepresentative training samples with simulated galaxies possessing otherwise unrepresented features. When we select simulated galaxies with ( g - z ) color, i -band magnitude, and redshift outside the range of the original training sample, we are able to reduce the outlier fraction of the photo- z estimates for simulated LSST data by nearly 50% and the normalized median absolute deviation (NMAD) by 56%. When compared to a fully representative training sample, augmentation can recover nearly 70% of the degradation in the outlier fraction and 80% of the degradation in NMAD. Training sample augmentation is a simple and effective way to improve training samples for photo- z 's without requiring additional spectroscopic samples.

View Journal Article

Sponsoring Organization:: USDOE

Grant/Contract Number:: AC02-76SF00515; SC0010008; SC0011665

OSTI ID:: 2349559

Journal Information:: The Astrophysical Journal. Letters, Journal Name: The Astrophysical Journal. Letters Journal Issue: 1 Vol. 967; ISSN 2041-8205

Publisher:: American Astronomical SocietyCopyright Statement

Country of Publication:: United States

Language:: English

References (42)

Conditional density estimation tools in python and R with applications to photometric redshifts and likelihood-free cosmological inference Dalmasso, N.; Pospisil, T.; Lee, A. B. Astronomy and Computing, Vol. 30 https://doi.org/10.1016/j.ascom.2019.100362	journal	January 2020
Galacticus: A semi-analytic model of galaxy formation Benson, Andrew J. New Astronomy, Vol. 17, Issue 2 https://doi.org/10.1016/j.newast.2011.07.004	journal	February 2012
KiDS-1000 catalogue: Redshift distributions and their calibration Hildebrandt, H.; van den Busch, J. L.; Wright, A. H. Astronomy & Astrophysics, Vol. 647 https://doi.org/10.1051/0004-6361/202039018	journal	March 2021
KiDS-1000 Cosmology: Multi-probe weak gravitational lensing and spectroscopic galaxy clustering constraints Heymans, Catherine; Tröster, Tilman; Asgari, Marika Astronomy & Astrophysics, Vol. 646 https://doi.org/10.1051/0004-6361/202039063	journal	February 2021
Euclid preparation No authors listed Astronomy & Astrophysics, Vol. 662 https://doi.org/10.1051/0004-6361/202141938	journal	June 2022
KiDS-1000: Cosmic shear with enhanced redshift calibration van den Busch, J. L.; Wright, A. H.; Hildebrandt, H. Astronomy & Astrophysics, Vol. 664 https://doi.org/10.1051/0004-6361/202142083	journal	August 2022
The Propagation of Uncertainties in Stellar Population Synthesis Modeling. i. the Relevance of Uncertain Aspects of Stellar Evolution and the Initial mass Function to the Derived Physical Properties of Galaxies Conroy, Charlie; Gunn, James E.; White, Martin The Astrophysical Journal, Vol. 699, Issue 1 https://doi.org/10.1088/0004-637X/699/1/486	journal	June 2009
Estimating Luminosities and Stellar Masses of Galaxies Photometrically Without Determining Redshifts Hsieh, B. C.; Yee, H. K. C. The Astrophysical Journal, Vol. 792, Issue 2 https://doi.org/10.1088/0004-637X/792/2/102	journal	August 2014
Mapping the Galaxy Color–Redshift Relation: Optimal Photometric Redshift Calibration Strategies for Cosmology Surveys Masters, Daniel; Capak, Peter; Stern, Daniel The Astrophysical Journal, Vol. 813, Issue 1 https://doi.org/10.1088/0004-637X/813/1/53	journal	October 2015
The Sensitivity of GPz Estimates of Photo-z Posterior PDFs to Realistically Complex Training Set Imperfections Stylianou, Natalia; Malz, Alex I.; Hatfield, Peter Publications of the Astronomical Society of the Pacific, Vol. 134, Issue 1034 https://doi.org/10.1088/1538-3873/ac59bf	journal	April 2022
Generating synthetic cosmological data with GalSampler Hearin, Andrew; Korytov, Danila; Kovacs, Eve Monthly Notices of the Royal Astronomical Society, Vol. 495, Issue 4 https://doi.org/10.1093/mnras/staa1495	journal	June 2020
Evaluation of probabilistic photometric redshift estimation approaches for The Rubin Observatory Legacy Survey of Space and Time (LSST) Schmidt, S. J.; Malz, A. I.; Soo, J. Y. H. Monthly Notices of the Royal Astronomical Society https://doi.org/10.1093/mnras/staa2799	journal	September 2020
Dark Energy Survey Year 3 results: redshift calibration of the weak lensing source galaxies Myles, J.; Alarcon, A.; Amon, A. Monthly Notices of the Royal Astronomical Society, Vol. 505, Issue 3 https://doi.org/10.1093/mnras/stab1515	journal	May 2021
Weak lensing tomographic redshift distribution inference for the Hyper Suprime-Cam Subaru Strategic Program three-year shape catalogue Rau, Markus Michael; Dalal, Roohi; Zhang, Tianqing Monthly Notices of the Royal Astronomical Society, Vol. 524, Issue 4 https://doi.org/10.1093/mnras/stad1962	journal	July 2023
TPZ: photometric redshift PDFs and ancillary information by using prediction trees and random forests Carrasco Kind, Matias; Brunner, Robert J. Monthly Notices of the Royal Astronomical Society, Vol. 432, Issue 2 https://doi.org/10.1093/mnras/stt574	journal	May 2013
On the realistic validation of photometric redshifts Beck, R.; Lin, C. -A.; Ishida, E. E. O. Monthly Notices of the Royal Astronomical Society, Vol. 468, Issue 4 https://doi.org/10.1093/mnras/stx687	journal	March 2017
UniverseMachine: The correlation between galaxy growth and dark matter halo assembly from z = 0−10 Behroozi, Peter; Wechsler, Risa H.; Hearin, Andrew P. Monthly Notices of the Royal Astronomical Society, Vol. 488, Issue 3 https://doi.org/10.1093/mnras/stz1182	journal	May 2019
Phenotypic redshifts with self-organizing maps: A novel method to characterize redshift distributions of source galaxies for weak lensing Buchs, R.; Davis, C.; Gruen, D. Monthly Notices of the Royal Astronomical Society, Vol. 489, Issue 1 https://doi.org/10.1093/mnras/stz2162	journal	August 2019
The Hyper Suprime-Cam SSP Survey: Overview and survey design Aihara, Hiroaki; Arimoto, Nobuo; Armstrong, Robert Publications of the Astronomical Society of Japan, Vol. 70, Issue SP1 https://doi.org/10.1093/pasj/psx066	journal	September 2017
Second data release of the Hyper Suprime-Cam Subaru Strategic Program Aihara, Hiroaki; AlSayyad, Yusra; Ando, Makoto Publications of the Astronomical Society of Japan https://doi.org/10.1093/pasj/psz103	journal	October 2019
Hyper Suprime-Cam Year 3 results: Cosmology from galaxy clustering and weak lensing with HSC and SDSS using the emulator based halo model Miyatake, Hironao; Sugiyama, Sunao; Takada, Masahiro Physical Review D, Vol. 108, Issue 12 https://doi.org/10.1103/PhysRevD.108.123517	journal	December 2023
Hyper Suprime-Cam Year 3 results: Cosmology from galaxy clustering and weak lensing with HSC and SDSS using the minimal bias model Sugiyama, Sunao; Miyatake, Hironao; More, Surhud Physical Review D, Vol. 108, Issue 12 https://doi.org/10.1103/PhysRevD.108.123521	journal	December 2023
Dark Energy Survey year 1 results: Cosmological constraints from galaxy clustering and weak lensing Abbott, T. M. C.; Abdalla, F. B.; Alarcon, A. Physical Review D, Vol. 98, Issue 4 https://doi.org/10.1103/physrevd.98.043526	journal	August 2018
Synthetic Biological Signals Machine-Generated by GPT-2 Improve the Classification of EEG and EMG Through Data Augmentation Bird, Jordan J.; Pritchard, Michael; Fratini, Antonio IEEE Robotics and Automation Letters, Vol. 6, Issue 2 https://doi.org/10.1109/LRA.2021.3056355	journal	April 2021
The cosmological simulation code gadget-2 Springel, Volker Monthly Notices of the Royal Astronomical Society, Vol. 364, Issue 4 https://doi.org/10.1111/j.1365-2966.2005.09655.x	journal	December 2005
The Dark Energy Spectroscopic Instrument (DESI) Flaugher, Brenna; Bebek, Chris SPIE Astronomical Telescopes + Instrumentation, SPIE Proceedings https://doi.org/10.1117/12.2057105	conference	July 2014
A survey on Image Data Augmentation for Deep Learning Shorten, Connor; Khoshgoftaar, Taghi M. Journal of Big Data, Vol. 6, Issue 1 https://doi.org/10.1186/s40537-019-0197-0	journal	July 2019
Converting high-dimensional regression to high-dimensional conditional density estimation Izbicki, Rafael; B. Lee, Ann Electronic Journal of Statistics, Vol. 11, Issue 2 https://doi.org/10.1214/17-EJS1302	journal	January 2017
4MOST: Project overview and information for the First Call for Proposals De Jong, Roelof S.; Agertz, Oscar; Berbel, Alex Agudo European Southern Observatory (ESO) https://doi.org/10.18727/0722-6691/5117	text	January 2019
The LSST-DESC 3x2pt Tomography Optimization Challenge Zuntz, Joe; Lanusse, François; Malz, Alex I. The Open Journal of Astrophysics, Vol. 4, Issue 1 https://doi.org/10.21105/astro.2108.13418	journal	October 2021
Augmentor: An Image Augmentation Library for Machine Learning D. Bloice, Marcus; Stocker, Christof; Holzinger, Andreas The Journal of Open Source Software, Vol. 2, Issue 19 https://doi.org/10.21105/joss.00432	journal	November 2017
LSST: From Science Drivers to Reference Design and Anticipated Data Products Ivezić, Željko; Kahn, Steven M.; Tyson, J. Anthony The Astrophysical Journal, Vol. 873, Issue 2 https://doi.org/10.3847/1538-4357/ab042c	journal	March 2019
Using a Neural Network Classifier to Select Galaxies with the Most Accurate Photometric Redshifts Broussard, Adam; Gawiser, Eric The Astrophysical Journal, Vol. 922, Issue 2 https://doi.org/10.3847/1538-4357/ac2147	journal	November 2021
ADDGALS: Simulated Sky Catalogs for Wide Field Galaxy Surveys Wechsler, Risa H.; DeRose, Joseph; Busha, Michael T. The Astrophysical Journal, Vol. 931, Issue 2 https://doi.org/10.3847/1538-4357/ac5b0a	journal	June 2022
Improved Tomographic Binning of 3 × 2 pt Lens Samples: Neural Network Classifiers and Optimal Bin Assignments Moskowitz, Irene; Gawiser, Eric; Bault, Abby The Astrophysical Journal, Vol. 950, Issue 1 https://doi.org/10.3847/1538-4357/accc88	journal	June 2023
Improving Photometric Redshift Estimation for Cosmology with LSST Using Bayesian Neural Networks Jones, Evan; Do, Tuan; Boscoe, Bernie The Astrophysical Journal, Vol. 964, Issue 2 https://doi.org/10.3847/1538-4357/ad2070	journal	March 2024
The Outer Rim Simulation: A Path to Many-core Supercomputers Heitmann, Katrin; Finkel, Hal; Pope, Adrian The Astrophysical Journal Supplement Series, Vol. 245, Issue 1 https://doi.org/10.3847/1538-4365/ab4da1	journal	November 2019
CosmoDC2: A Synthetic Sky Catalog for Dark Energy Science with LSST Korytov, Danila; Hearin, Andrew; Kovacs, Eve The Astrophysical Journal Supplement Series, Vol. 245, Issue 2 https://doi.org/10.3847/1538-4365/ab510c	journal	December 2019
The LSST DESC DC2 Simulated Sky Survey Abolfathi, Bela; Alonso, David; Armstrong, Robert The Astrophysical Journal Supplement Series, Vol. 253, Issue 1 https://doi.org/10.3847/1538-4365/abd62c	journal	March 2021
COSMOS2020: A Panchromatic View of the Universe to z ∼ 10 from Two Complementary Catalogs Weaver, J. R.; Kauffmann, O. B.; Ilbert, O. The Astrophysical Journal Supplement Series, Vol. 258, Issue 1 https://doi.org/10.3847/1538-4365/ac3078	journal	January 2022
jfcrenshaw/pzflow: v3.1.1 Crenshaw, John Franklin; Yan, Zi'ang; Doster, Vladislav https://doi.org/10.5281/zenodo.7843901	software	April 2023
LSSTDESC/RAIL: v0.98.5 Schmidt, Sam; Gschwend, Julia; Crenshaw, John Franklin https://doi.org/10.5281/zenodo.7927358	software	May 2023

Similar Records

Photometric Redshifts with the LSST. II. The Impact of Near-infrared and Near-ultraviolet Photometry

Journal Article · Mon Jun 01 00:00:00 EDT 2020 · The Astronomical Journal (Online) · OSTI ID:23013228

Photometric Redshifts with the LSST. II. The Impact of Near-infrared and Near-ultraviolet Photometry

Journal Article · Tue May 12 20:00:00 EDT 2020 · The Astronomical Journal (Online) · OSTI ID:1802007

Using a Neural Network Classifier to Select Galaxies with the Most Accurate Photometric Redshifts

Journal Article · Thu Nov 25 19:00:00 EST 2021 · The Astrophysical Journal · OSTI ID:1981229

Improving Photometric Redshift Estimates with Training Sample Augmentation

Citation Formats

References (42)

Similar Records

Related Subjects