Echoes in the Noise: Posterior Samples of Faint Galaxy Surface Brightness Profiles with Score-Based Likelihoods and Priors

Alexandre Adam; Connor Stone; Connor Bottrell; Ronan Legin; Yashar Hezaveh; Laurence Perreault-Levasseur

doi:10.48550/arXiv.2311.18002

Echoes in the Noise: Posterior Samples of Faint Galaxy Surface Brightness Profiles with Score-Based Likelihoods and Priors

Alexandre Adam, Connor Stone, Connor Bottrell, Ronan Legin, Yashar Hezaveh, Laurence Perreault-Levasseur

International Centre for Radio Astronomy Research (ICRAR)

Research output: Working paper › Preprint

21 Downloads (Pure)

Abstract

Examining the detailed structure of galaxy populations provides valuable insights into their formation and evolution mechanisms. Significant barriers to such analysis are the non-trivial noise properties of real astronomical images and the point spread function (PSF) which blurs structure. Here we present a framework which combines recent advances in score-based likelihood characterization and diffusion model priors to perform a Bayesian analysis of image deconvolution. The method, when applied to minimally processed \emph{Hubble Space Telescope} (\emph{HST}) data, recovers structures which have otherwise only become visible in next-generation \emph{James Webb Space Telescope} (\emph{JWST}) imaging.

Original language	English
Place of Publication	USA
Publisher	arXiv
DOIs	https://doi.org/10.48550/arXiv.2311.18002
Publication status	Published - 29 Nov 2023

Access to Document

10.48550/arXiv.2311.18002Licence: CC BY

2311.18002v1

Cite this

@techreport{5d4e94fd23e04fd5bb86468b0e6e55e8,

title = "Echoes in the Noise: Posterior Samples of Faint Galaxy Surface Brightness Profiles with Score-Based Likelihoods and Priors",

abstract = " Examining the detailed structure of galaxy populations provides valuable insights into their formation and evolution mechanisms. Significant barriers to such analysis are the non-trivial noise properties of real astronomical images and the point spread function (PSF) which blurs structure. Here we present a framework which combines recent advances in score-based likelihood characterization and diffusion model priors to perform a Bayesian analysis of image deconvolution. The method, when applied to minimally processed \emph{Hubble Space Telescope} (\emph{HST}) data, recovers structures which have otherwise only become visible in next-generation \emph{James Webb Space Telescope} (\emph{JWST}) imaging. ",

keywords = "astro-ph.IM, cs.CV",

author = "Alexandre Adam and Connor Stone and Connor Bottrell and Ronan Legin and Yashar Hezaveh and Laurence Perreault-Levasseur",

note = "5+5 pages, 10 figures, Machine Learning and the Physical Sciences Workshop, NeurIPS 2023",

year = "2023",

month = nov,

day = "29",

doi = "10.48550/arXiv.2311.18002",

language = "English",

publisher = "arXiv",

address = "United States",

type = "WorkingPaper",

institution = "arXiv",

}

TY - UNPB

T1 - Echoes in the Noise

T2 - Posterior Samples of Faint Galaxy Surface Brightness Profiles with Score-Based Likelihoods and Priors

AU - Adam, Alexandre

AU - Stone, Connor

AU - Bottrell, Connor

AU - Legin, Ronan

AU - Hezaveh, Yashar

AU - Perreault-Levasseur, Laurence

N1 - 5+5 pages, 10 figures, Machine Learning and the Physical Sciences Workshop, NeurIPS 2023

PY - 2023/11/29

Y1 - 2023/11/29

N2 - Examining the detailed structure of galaxy populations provides valuable insights into their formation and evolution mechanisms. Significant barriers to such analysis are the non-trivial noise properties of real astronomical images and the point spread function (PSF) which blurs structure. Here we present a framework which combines recent advances in score-based likelihood characterization and diffusion model priors to perform a Bayesian analysis of image deconvolution. The method, when applied to minimally processed \emph{Hubble Space Telescope} (\emph{HST}) data, recovers structures which have otherwise only become visible in next-generation \emph{James Webb Space Telescope} (\emph{JWST}) imaging.

AB - Examining the detailed structure of galaxy populations provides valuable insights into their formation and evolution mechanisms. Significant barriers to such analysis are the non-trivial noise properties of real astronomical images and the point spread function (PSF) which blurs structure. Here we present a framework which combines recent advances in score-based likelihood characterization and diffusion model priors to perform a Bayesian analysis of image deconvolution. The method, when applied to minimally processed \emph{Hubble Space Telescope} (\emph{HST}) data, recovers structures which have otherwise only become visible in next-generation \emph{James Webb Space Telescope} (\emph{JWST}) imaging.

KW - astro-ph.IM

KW - cs.CV

U2 - 10.48550/arXiv.2311.18002

DO - 10.48550/arXiv.2311.18002

M3 - Preprint

BT - Echoes in the Noise

PB - arXiv

CY - USA

ER -

Echoes in the Noise: Posterior Samples of Faint Galaxy Surface Brightness Profiles with Score-Based Likelihoods and Priors

Abstract

Access to Document

Fingerprint

Cite this