Accelerated, scalable and reproducible AI-driven gravitational wave detection

E. A. Huerta; Asad Khan; Xiaobo Huang; Minyang Tian; Maksim Levental; Ryan Chard; Wei Wei; Maeve Heflin; Daniel S. Katz; Volodymyr Kindratenko; Dawei Mu; Ben Blaiszik; Ian Foster

doi:10.1038/s41550-021-01405-0

Accelerated, scalable and reproducible AI-driven gravitational wave detection

E. A. Huerta, Asad Khan, Xiaobo Huang, Minyang Tian, Maksim Levental, Ryan Chard, Wei Wei, Maeve Heflin, Daniel S. Katz, Volodymyr Kindratenko, Dawei Mu, Ben Blaiszik, Ian Foster

Research output: Contribution to journal › Article › peer-review

Abstract

The development of reusable artificial intelligence (AI) models for wider use and rigorous validation by the community promises to unlock new opportunities in multi-messenger astrophysics. Here we develop a workflow that connects the Data and Learning Hub for Science, a repository for publishing AI models, with the Hardware-Accelerated Learning (HAL) cluster, using funcX as a universal distributed computing service. Using this workflow, an ensemble of four openly available AI models can be run on HAL to process an entire month’s worth (August 2017) of advanced Laser Interferometer Gravitational-Wave Observatory data in just seven minutes, identifying all four binary black hole mergers previously identified in this dataset and reporting no misclassifications. This approach combines advances in AI, distributed computing and scientific data infrastructure to open new pathways to conduct reproducible, accelerated, data-driven discovery.

Original language	English (US)
Pages (from-to)	1062-1068
Number of pages	7
Journal	Nature Astronomy
Volume	5
Issue number	10
DOIs	https://doi.org/10.1038/s41550-021-01405-0
State	Published - Oct 2021

ASJC Scopus subject areas

Astronomy and Astrophysics

Online availability

10.1038/s41550-021-01405-0

Library availability

Discover UIUC Full Text

Cite this

@article{b21be5c4588343829262fe070800620b,

title = "Accelerated, scalable and reproducible AI-driven gravitational wave detection",

abstract = "The development of reusable artificial intelligence (AI) models for wider use and rigorous validation by the community promises to unlock new opportunities in multi-messenger astrophysics. Here we develop a workflow that connects the Data and Learning Hub for Science, a repository for publishing AI models, with the Hardware-Accelerated Learning (HAL) cluster, using funcX as a universal distributed computing service. Using this workflow, an ensemble of four openly available AI models can be run on HAL to process an entire month{\textquoteright}s worth (August 2017) of advanced Laser Interferometer Gravitational-Wave Observatory data in just seven minutes, identifying all four binary black hole mergers previously identified in this dataset and reporting no misclassifications. This approach combines advances in AI, distributed computing and scientific data infrastructure to open new pathways to conduct reproducible, accelerated, data-driven discovery.",

author = "Huerta, {E. A.} and Asad Khan and Xiaobo Huang and Minyang Tian and Maksim Levental and Ryan Chard and Wei Wei and Maeve Heflin and Katz, {Daniel S.} and Volodymyr Kindratenko and Dawei Mu and Ben Blaiszik and Ian Foster",

note = "Publisher Copyright: {\textcopyright} 2021, The Author(s), under exclusive licence to Springer Nature Limited.",

year = "2021",

month = oct,

doi = "10.1038/s41550-021-01405-0",

language = "English (US)",

volume = "5",

pages = "1062--1068",

journal = "Nature Astronomy",

issn = "2397-3366",

publisher = "Nature Publishing Group",

number = "10",

}

TY - JOUR

T1 - Accelerated, scalable and reproducible AI-driven gravitational wave detection

AU - Huerta, E. A.

AU - Khan, Asad

AU - Huang, Xiaobo

AU - Tian, Minyang

AU - Levental, Maksim

AU - Chard, Ryan

AU - Wei, Wei

AU - Heflin, Maeve

AU - Katz, Daniel S.

AU - Kindratenko, Volodymyr

AU - Mu, Dawei

AU - Blaiszik, Ben

AU - Foster, Ian

PY - 2021/10

Y1 - 2021/10

N2 - The development of reusable artificial intelligence (AI) models for wider use and rigorous validation by the community promises to unlock new opportunities in multi-messenger astrophysics. Here we develop a workflow that connects the Data and Learning Hub for Science, a repository for publishing AI models, with the Hardware-Accelerated Learning (HAL) cluster, using funcX as a universal distributed computing service. Using this workflow, an ensemble of four openly available AI models can be run on HAL to process an entire month’s worth (August 2017) of advanced Laser Interferometer Gravitational-Wave Observatory data in just seven minutes, identifying all four binary black hole mergers previously identified in this dataset and reporting no misclassifications. This approach combines advances in AI, distributed computing and scientific data infrastructure to open new pathways to conduct reproducible, accelerated, data-driven discovery.

AB - The development of reusable artificial intelligence (AI) models for wider use and rigorous validation by the community promises to unlock new opportunities in multi-messenger astrophysics. Here we develop a workflow that connects the Data and Learning Hub for Science, a repository for publishing AI models, with the Hardware-Accelerated Learning (HAL) cluster, using funcX as a universal distributed computing service. Using this workflow, an ensemble of four openly available AI models can be run on HAL to process an entire month’s worth (August 2017) of advanced Laser Interferometer Gravitational-Wave Observatory data in just seven minutes, identifying all four binary black hole mergers previously identified in this dataset and reporting no misclassifications. This approach combines advances in AI, distributed computing and scientific data infrastructure to open new pathways to conduct reproducible, accelerated, data-driven discovery.

UR - http://www.scopus.com/inward/record.url?scp=85109256677&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85109256677&partnerID=8YFLogxK

U2 - 10.1038/s41550-021-01405-0

DO - 10.1038/s41550-021-01405-0

M3 - Article

AN - SCOPUS:85109256677

SN - 2397-3366

VL - 5

SP - 1062

EP - 1068

JO - Nature Astronomy

JF - Nature Astronomy

IS - 10

ER -

Accelerated, scalable and reproducible AI-driven gravitational wave detection

Abstract

ASJC Scopus subject areas

Online availability

Library availability

Related links

Fingerprint

Cite this