Reinforcement Learning for Data-driven Workflows in Radio Interferometry. I. Principal Demonstration in Calibration

Radio interferometry is an observational technique used to study astrophysical phenomena. Data gathered by an interferometer require substantial processing before astronomers can extract scientific information from them. Data processing consists of a sequence of calibration and analysis procedures w...

Full description

Saved in:
Bibliographic Details
Main Authors: Brian M. Kirk, Urvashi Rau, Ramyaa Ramyaa
Format: Article
Language:English
Published: IOP Publishing 2024-01-01
Series:The Astronomical Journal
Subjects:
Online Access:https://doi.org/10.3847/1538-3881/ad88f6
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850122840175542272
author Brian M. Kirk
Urvashi Rau
Ramyaa Ramyaa
author_facet Brian M. Kirk
Urvashi Rau
Ramyaa Ramyaa
author_sort Brian M. Kirk
collection DOAJ
description Radio interferometry is an observational technique used to study astrophysical phenomena. Data gathered by an interferometer require substantial processing before astronomers can extract scientific information from them. Data processing consists of a sequence of calibration and analysis procedures where choices must be made about the sequence of procedures as well as the specific configuration of the procedure itself. These choices are typically based on a combination of measurable data characteristics, an understanding of the instrument itself, an appreciation of the trade-offs between compute cost and accuracy, and a learned understanding of what is considered best practice. A metric of absolute correctness is not always available and validity is often subject to human judgment. The underlying principles and software configurations to discern a reasonable workflow for a given data set is the subject of training workshops for students and scientists. Our goal is to use objective metrics that quantify best practice, and numerically map out the decision space with respect to our metrics. With these objective metrics we demonstrate an automated, data-driven, decision system that is capable of sequencing the optimal action(s) for processing interferometric data. This paper introduces a simplified description of the principles behind interferometry and the procedures required for data processing. We highlight the issues with current automation approaches and propose our ideas for solving these bottlenecks. A prototype is demonstrated and the results are discussed.
format Article
id doaj-art-338fabd9bb0a4cce80878cf19ab67644
institution OA Journals
issn 1538-3881
language English
publishDate 2024-01-01
publisher IOP Publishing
record_format Article
series The Astronomical Journal
spelling doaj-art-338fabd9bb0a4cce80878cf19ab676442025-08-20T02:34:44ZengIOP PublishingThe Astronomical Journal1538-38812024-01-0116914310.3847/1538-3881/ad88f6Reinforcement Learning for Data-driven Workflows in Radio Interferometry. I. Principal Demonstration in CalibrationBrian M. Kirk0https://orcid.org/0000-0002-7753-9130Urvashi Rau1Ramyaa Ramyaa2New Mexico Institute of Mining and Technology , USA; National Radio Astronomy Observatory , USANational Radio Astronomy Observatory , USANew Mexico Institute of Mining and Technology , USARadio interferometry is an observational technique used to study astrophysical phenomena. Data gathered by an interferometer require substantial processing before astronomers can extract scientific information from them. Data processing consists of a sequence of calibration and analysis procedures where choices must be made about the sequence of procedures as well as the specific configuration of the procedure itself. These choices are typically based on a combination of measurable data characteristics, an understanding of the instrument itself, an appreciation of the trade-offs between compute cost and accuracy, and a learned understanding of what is considered best practice. A metric of absolute correctness is not always available and validity is often subject to human judgment. The underlying principles and software configurations to discern a reasonable workflow for a given data set is the subject of training workshops for students and scientists. Our goal is to use objective metrics that quantify best practice, and numerically map out the decision space with respect to our metrics. With these objective metrics we demonstrate an automated, data-driven, decision system that is capable of sequencing the optimal action(s) for processing interferometric data. This paper introduces a simplified description of the principles behind interferometry and the procedures required for data processing. We highlight the issues with current automation approaches and propose our ideas for solving these bottlenecks. A prototype is demonstrated and the results are discussed.https://doi.org/10.3847/1538-3881/ad88f6Computational methodsComputational astronomyInterdisciplinary astronomyRadio astronomy
spellingShingle Brian M. Kirk
Urvashi Rau
Ramyaa Ramyaa
Reinforcement Learning for Data-driven Workflows in Radio Interferometry. I. Principal Demonstration in Calibration
The Astronomical Journal
Computational methods
Computational astronomy
Interdisciplinary astronomy
Radio astronomy
title Reinforcement Learning for Data-driven Workflows in Radio Interferometry. I. Principal Demonstration in Calibration
title_full Reinforcement Learning for Data-driven Workflows in Radio Interferometry. I. Principal Demonstration in Calibration
title_fullStr Reinforcement Learning for Data-driven Workflows in Radio Interferometry. I. Principal Demonstration in Calibration
title_full_unstemmed Reinforcement Learning for Data-driven Workflows in Radio Interferometry. I. Principal Demonstration in Calibration
title_short Reinforcement Learning for Data-driven Workflows in Radio Interferometry. I. Principal Demonstration in Calibration
title_sort reinforcement learning for data driven workflows in radio interferometry i principal demonstration in calibration
topic Computational methods
Computational astronomy
Interdisciplinary astronomy
Radio astronomy
url https://doi.org/10.3847/1538-3881/ad88f6
work_keys_str_mv AT brianmkirk reinforcementlearningfordatadrivenworkflowsinradiointerferometryiprincipaldemonstrationincalibration
AT urvashirau reinforcementlearningfordatadrivenworkflowsinradiointerferometryiprincipaldemonstrationincalibration
AT ramyaaramyaa reinforcementlearningfordatadrivenworkflowsinradiointerferometryiprincipaldemonstrationincalibration