descriptionPublicationkeyboard_double_arrow_right Article , Other literature type , Preprint 21 Jan 2021Embargo end date: 01 Jan 2020 English Publisher:Oxford University Press (OUP)Journal:Monthly Notices of the Royal Astronomical Society, volume 502, pages 2,770-2,786 (issn: 0035-8711, eissn: 1365-2966,

Authors: Sunil Mucesh; W. G. Hartley; A. Palmese; O. Lahav; L. Whiteway; Asa F. L. Bluck; A. Alarcón; +68 Authors

doi: 10.1093/mnras/stab164 , 10.60692/9w9f6-erg95 , 10.48550/arxiv.2012.05928 , 10.60692/pzpz0-k7078

arXiv: 2012.05928

handle: 10261/262803 , 11368/2988341 , 10852/89328

A machine learning approach to galaxy properties: joint redshift–stellar mass probability distributions with Random Forest

- Summary
- Subjects
- Related research
  (1)
- Metrics

Abstract

ABSTRACTWe demonstrate that highly accurate joint redshift–stellar mass probability distribution functions (PDFs) can be obtained using the Random Forest (RF) machine learning (ML) algorithm, even with few photometric bands available. As an example, we use the Dark Energy Survey (DES), combined with the COSMOS2015 catalogue for redshifts and stellar masses. We build two ML models: one containing deep photometry in the griz bands, and the second reflecting the photometric scatter present in the main DES survey, with carefully constructed representative training data in each case. We validate our joint PDFs for 10 699 test galaxies by utilizing the copula probability integral transform and the Kendall distribution function, and their univariate counterparts to validate the marginals. Benchmarked against a basic set-up of the template-fitting code bagpipes, our ML-based method outperforms template fitting on all of our predefined performance metrics. In addition to accuracy, the RF is extremely fast, able to compute joint PDFs for a million galaxies in just under 6 min with consumer computer hardware. Such speed enables PDFs to be derived in real time within analysis codes, solving potential storage issues. As part of this work we have developed galpro1, a highly intuitive and efficient python package to rapidly generate multivariate PDFs on-the-fly. galpro is documented and available for researchers to use in their cosmology and galaxy evolution studies.

Related Organizations

Australian National University
Australia
University of Paris
France
Federal University of Rio de Janeiro
Brazil
University of Illinois at Urbana Champaign
United States
Sorbonne University
France

View all View all

Keywords

FOS: Computer and information sciences, statistical [Methods], Computer Science - Machine Learning, Artificial intelligence, software: data analysis, Astrophysics, Machine Learning (cs.LG), Gaussian Processes in Machine Learning, Probability distribution, Software: data analysis, Sparse Regression, Probability density function, Astrophysics - Cosmology and Nongalactic Astrophysic, data analysis [Methods], galaxies: fundamental parameter, Methods: statistical, Ecology, Galaxy Formation and Evolution in the Universe, Physics, Star formation, Python (programming language), Statistics, Galaxies: evolution, galaxies: fundamental parameters, Remote Sensing in Vegetation Monitoring and Phenology, 520, Cosmology, Algorithm, Software: public realese, fundamental parameters [Galaxies], Physical Sciences, Photometric redshift, software: data analysi, galaxies: evolution, Astrophysics - Instrumentation and Methods for Astrophysics, methods: data analysi, Astrophysics - Cosmology and Nongalactic Astrophysics, public realese [Software], Galaxies: fundamental parameters, Cosmology and Nongalactic Astrophysics (astro-ph.CO), methods: data analysis; methods: statistical; galaxies: evolution; galaxies: fundamental parameters; software: data analysis; software: public release; Astrophysics - Astrophysics of Galaxies; Astrophysics - Cosmology and Nongalactic Astrophysics; Astrophysics - Instrumentation and Methods for Astrophysics; Computer Science - Machine Learning, FOS: Physical sciences, [INFO] Computer Science [cs], Joint probability distribution, software: public release, Astrophysics - Astrophysics of Galaxie, Methods: data analysis, Artificial Intelligence, Stellar mass, FOS: Mathematics, Instrumentation and Methods for Astrophysics (astro-ph.IM), methods: statistical, 500, Astronomy and Astrophysics, Redshift, evolution [Galaxies], methods: data analysis, Astrophysics - Astrophysics of Galaxies, Computer science, Stars, Galaxy, Operating system, Physics and Astronomy, [PHYS.PHYS.PHYS-INS-DET] Physics [physics]/Physics [physics]/Instrumentation and Detectors [physics.ins-det], data analysis [Software], Photometry (optics), Astrophysics of Galaxies (astro-ph.GA), FOS: Biological sciences, Environmental Science, Computer Science, [PHYS.ASTR] Physics [physics]/Astrophysics [astro-ph], Astrophysics - Instrumentation and Methods for Astrophysic, Mathematics, Random forest

1 Research products, page 1 of 1

galpro software on GitHub
IsRelatedTo

Impact byBIP!

	selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).	36
	popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.	Top 10%
	influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).	Average
	impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.	Top 10%