Photometric redshifts for S-PLUS using machine learning techniques
Autor(a) principal: | |
---|---|
Data de Publicação: | 2019 |
Tipo de documento: | Dissertação |
Idioma: | eng |
Título da fonte: | Biblioteca Digital de Teses e Dissertações da USP |
Texto Completo: | https://www.teses.usp.br/teses/disponiveis/14/14131/tde-14082020-174722/ |
Resumo: | This work focuses on the obtention of photometric redshifts of galaxies using the machine learning codes ANNz2, GPz, and Deep Learning models made with Keras. We take advantage of the great opportunity that the new multiband survey of the austral sky, called Southern Photometric Local Universe Survey (S-PLUS), provides due to the adoption of an unique filter system, composed of five broad-band filters and seven narrow-band filters. Besides the use of magnitudes, it is also possible to use non-photometric features with machine learning methods, such as object sizes, their full width at half maximum and their maximum surface brightness, in order to improve the results. This work used data from the S-PLUS Data Release 1, together with two other large projects, the Sloan Digital Sky Survey (SDSS) Data Release 15, and the unWISE catalogue of the Wide-field Infrared Survey Explorer (WISE), in the Stripe-82 region. Amongst the three algorithms compared in this work, the one which performs better overall is the deep-learning based method. The photometric redshifts obtained with this code have precision of 2.49% for galaxies with r-band magnitude between 16 and 21, bias of 0.4% and outlier fraction equal to 0.64%. When compared to the currently used method for photometric redshift determination in S-PLUS, the template fitting code BPZ, it is noticed that machine learning methods have higher accuracy, less bias and lower outlier fraction. An analysis regarding the probability distribution function is made, concluding that the machine learning algorithms present broader distributions when compared to the BPZ code. |
id |
USP_b73749e78051121a3786bc9850267835 |
---|---|
oai_identifier_str |
oai:teses.usp.br:tde-14082020-174722 |
network_acronym_str |
USP |
network_name_str |
Biblioteca Digital de Teses e Dissertações da USP |
repository_id_str |
2721 |
spelling |
Photometric redshifts for S-PLUS using machine learning techniquesRedshifts fotométricos para o S-PLUS utilizando técnicas de aprendizado de máquina.Aprendizado de máquinaGaláxiasGalaxiesGalaxy surveysLevantamentos de galáxias.Machine learningPhotometric redshiftRedshift fotométricoThis work focuses on the obtention of photometric redshifts of galaxies using the machine learning codes ANNz2, GPz, and Deep Learning models made with Keras. We take advantage of the great opportunity that the new multiband survey of the austral sky, called Southern Photometric Local Universe Survey (S-PLUS), provides due to the adoption of an unique filter system, composed of five broad-band filters and seven narrow-band filters. Besides the use of magnitudes, it is also possible to use non-photometric features with machine learning methods, such as object sizes, their full width at half maximum and their maximum surface brightness, in order to improve the results. This work used data from the S-PLUS Data Release 1, together with two other large projects, the Sloan Digital Sky Survey (SDSS) Data Release 15, and the unWISE catalogue of the Wide-field Infrared Survey Explorer (WISE), in the Stripe-82 region. Amongst the three algorithms compared in this work, the one which performs better overall is the deep-learning based method. The photometric redshifts obtained with this code have precision of 2.49% for galaxies with r-band magnitude between 16 and 21, bias of 0.4% and outlier fraction equal to 0.64%. When compared to the currently used method for photometric redshift determination in S-PLUS, the template fitting code BPZ, it is noticed that machine learning methods have higher accuracy, less bias and lower outlier fraction. An analysis regarding the probability distribution function is made, concluding that the machine learning algorithms present broader distributions when compared to the BPZ code.O foco deste trabalho é a obtenção de redshifts fotométricos de galáxias utilizando os códigos de aprendizado de máquina ANNz2, GPz e modelos de aprendizado profundo feitos com o Keras. Nós aproveitamos a excelente oportunidade que o novo mapeamento multicor do céu austral, chamado Southern Photometric Local Universe Survey (S-PLUS), oferece ao utilizar um sistema único de filtros, composto por cinco filtros largos e sete filtros estreitos. Além do uso das magnitudes, também é possível utilizar características não fotométricas em métodos de aprendizado de máquina, como o tamanho dos objetos, sua largura à meia altura e seu brilho superficial, de forma a melhorar os resultados. Para este trabalho, foram usados dados provenientes do Data Release 1 do S-PLUS, unido à outros dois grandes projetos, o Data Release 15 do Sloan Digital Sky Survey (SDSS) e o catálogo unWISE do Wide-field Infrared Survey Explorer (WISE), na região da Stripe-82. Dentre os três algoritmos comparados neste trabalho, o que apresentou a melhor performance geral foi o baseado em aprendizado profundo. Os redshifts fotométricos obtidos com este método têm precisão de 2.49% para galáxias com magnitude r entre 16 e 21, com viés igual a 0.4% e fração de outliers de 0.64%. Em comparação com o método utilizado atualmente para a estimativa de redshifts fotométricos no S-PLUS, o código de ajuste de templates BPZ, foi constatado que os métodos de aprendizado de máquina têm precisão superior, viés inferior e menor fração de outliers. Uma análise das funções de distribuição de probabilidades é feita, concluindo-se que os algoritmos de aprendizado de máquina apresentam distribuições mais largas quando comparadas às do código BPZ.Biblioteca Digitais de Teses e Dissertações da USPSodre Junior, LaerteLima, Erik Vinicius Rodrigues de2019-10-29info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttps://www.teses.usp.br/teses/disponiveis/14/14131/tde-14082020-174722/reponame:Biblioteca Digital de Teses e Dissertações da USPinstname:Universidade de São Paulo (USP)instacron:USPLiberar o conteúdo para acesso público.info:eu-repo/semantics/openAccesseng2020-08-18T21:58:59Zoai:teses.usp.br:tde-14082020-174722Biblioteca Digital de Teses e Dissertaçõeshttp://www.teses.usp.br/PUBhttp://www.teses.usp.br/cgi-bin/mtd2br.plvirginia@if.usp.br|| atendimento@aguia.usp.br||virginia@if.usp.bropendoar:27212020-08-18T21:58:59Biblioteca Digital de Teses e Dissertações da USP - Universidade de São Paulo (USP)false |
dc.title.none.fl_str_mv |
Photometric redshifts for S-PLUS using machine learning techniques Redshifts fotométricos para o S-PLUS utilizando técnicas de aprendizado de máquina. |
title |
Photometric redshifts for S-PLUS using machine learning techniques |
spellingShingle |
Photometric redshifts for S-PLUS using machine learning techniques Lima, Erik Vinicius Rodrigues de Aprendizado de máquina Galáxias Galaxies Galaxy surveys Levantamentos de galáxias. Machine learning Photometric redshift Redshift fotométrico |
title_short |
Photometric redshifts for S-PLUS using machine learning techniques |
title_full |
Photometric redshifts for S-PLUS using machine learning techniques |
title_fullStr |
Photometric redshifts for S-PLUS using machine learning techniques |
title_full_unstemmed |
Photometric redshifts for S-PLUS using machine learning techniques |
title_sort |
Photometric redshifts for S-PLUS using machine learning techniques |
author |
Lima, Erik Vinicius Rodrigues de |
author_facet |
Lima, Erik Vinicius Rodrigues de |
author_role |
author |
dc.contributor.none.fl_str_mv |
Sodre Junior, Laerte |
dc.contributor.author.fl_str_mv |
Lima, Erik Vinicius Rodrigues de |
dc.subject.por.fl_str_mv |
Aprendizado de máquina Galáxias Galaxies Galaxy surveys Levantamentos de galáxias. Machine learning Photometric redshift Redshift fotométrico |
topic |
Aprendizado de máquina Galáxias Galaxies Galaxy surveys Levantamentos de galáxias. Machine learning Photometric redshift Redshift fotométrico |
description |
This work focuses on the obtention of photometric redshifts of galaxies using the machine learning codes ANNz2, GPz, and Deep Learning models made with Keras. We take advantage of the great opportunity that the new multiband survey of the austral sky, called Southern Photometric Local Universe Survey (S-PLUS), provides due to the adoption of an unique filter system, composed of five broad-band filters and seven narrow-band filters. Besides the use of magnitudes, it is also possible to use non-photometric features with machine learning methods, such as object sizes, their full width at half maximum and their maximum surface brightness, in order to improve the results. This work used data from the S-PLUS Data Release 1, together with two other large projects, the Sloan Digital Sky Survey (SDSS) Data Release 15, and the unWISE catalogue of the Wide-field Infrared Survey Explorer (WISE), in the Stripe-82 region. Amongst the three algorithms compared in this work, the one which performs better overall is the deep-learning based method. The photometric redshifts obtained with this code have precision of 2.49% for galaxies with r-band magnitude between 16 and 21, bias of 0.4% and outlier fraction equal to 0.64%. When compared to the currently used method for photometric redshift determination in S-PLUS, the template fitting code BPZ, it is noticed that machine learning methods have higher accuracy, less bias and lower outlier fraction. An analysis regarding the probability distribution function is made, concluding that the machine learning algorithms present broader distributions when compared to the BPZ code. |
publishDate |
2019 |
dc.date.none.fl_str_mv |
2019-10-29 |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
format |
masterThesis |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
https://www.teses.usp.br/teses/disponiveis/14/14131/tde-14082020-174722/ |
url |
https://www.teses.usp.br/teses/disponiveis/14/14131/tde-14082020-174722/ |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
|
dc.rights.driver.fl_str_mv |
Liberar o conteúdo para acesso público. info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
Liberar o conteúdo para acesso público. |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.coverage.none.fl_str_mv |
|
dc.publisher.none.fl_str_mv |
Biblioteca Digitais de Teses e Dissertações da USP |
publisher.none.fl_str_mv |
Biblioteca Digitais de Teses e Dissertações da USP |
dc.source.none.fl_str_mv |
reponame:Biblioteca Digital de Teses e Dissertações da USP instname:Universidade de São Paulo (USP) instacron:USP |
instname_str |
Universidade de São Paulo (USP) |
instacron_str |
USP |
institution |
USP |
reponame_str |
Biblioteca Digital de Teses e Dissertações da USP |
collection |
Biblioteca Digital de Teses e Dissertações da USP |
repository.name.fl_str_mv |
Biblioteca Digital de Teses e Dissertações da USP - Universidade de São Paulo (USP) |
repository.mail.fl_str_mv |
virginia@if.usp.br|| atendimento@aguia.usp.br||virginia@if.usp.br |
_version_ |
1809090717498212352 |