CaltechAUTHORS
Published April 15, 2019 | Version Submitted + Published
Journal Article Open

Classifying the unknown: Discovering novel gravitational-wave detector glitches using similarity learning

Creators

  • 1. ROR icon Cardiff University
  • 2. ROR icon Northwestern University
  • 3. ROR icon California State University, Fullerton
  • 4. ROR icon Syracuse University
  • 5. ROR icon University of Illinois at Chicago
  • 6. ROR icon Adler Planetarium
  • 7. ROR icon California Institute of Technology
  • 8. ROR icon University of Portsmouth

Abstract

The observation of gravitational waves from compact binary coalescences by LIGO and Virgo has begun a new era in astronomy. A critical challenge in making detections is determining whether loud transient features in the data are caused by gravitational waves or by instrumental or environmental sources. The citizen-science project Gravity Spy has been demonstrated as an efficient infrastructure for classifying known types of noise transients (glitches) through a combination of data analysis performed by both citizen volunteers and machine learning. We present the next iteration of this project, using similarity indices to empower citizen scientists to create large data sets of unknown transients, which can then be used to facilitate supervised machine-learning characterization. This new evolution aims to alleviate a persistent challenge that plagues both citizen-science and instrumental detector work: the ability to build large samples of relatively rare events. Using two families of transient noise that appeared unexpectedly during LIGO's second observing run, we demonstrate the impact that the similarity indices could have had on finding these new glitch types in the Gravity Spy program.

Additional Information

© 2019 American Physical Society. Received 10 March 2019; published 16 April 2019. First, and foremost, we thank the many Gravity Spy participants that make this work possible. We thank Eliu Huerta, Alex Urban, and Patrick Sutton for their useful comments. Gravity Spy is partly supported by the National Science Foundation award INSPIRE 15-47880. O. P. is supported by NSF award AST-1559694. M. C. is supported by the David and Ellen Lee Postdoctoral Fellowship at the California Institute of Technology. C. P. .L. B. is supported by the CIERA Board of Visitors Research Professorship. In addition, computing was provided by the LIGO Data Grid which is supported by the National Science Foundation Grants PHY-0757058 and PHY-0823459. This work also used computing resources at CIERA funded by NSF PHY-1126812. This paper has been assigned LIGO document number LIGO-P1800352.

Attached Files

Published - PhysRevD.99.082002.pdf

Submitted - 1903.04058.pdf

Files

1903.04058.pdf

Files (4.3 MB)

Name Size
md5:98185c11032502104eb5e340caf69980
2.6 MB Preview Download
md5:b300968227849b159d07a26e365113b2
1.7 MB Preview Download

Additional details

Identifiers

Eprint ID
94738
Resolver ID
CaltechAUTHORS:20190417-084807411

Related works

Describes
https://arxiv.org/abs/1903.04058 (URL)

Funding

NSF
15-47880
NSF
AST-1559694
David and Ellen Lee Postdoctoral Scholarship
CIERA Board of Visitors Research Professorship
NSF
PHY-0757058
NSF
PHY-0823459
NSF
PHY-1126812

Dates

Created
2019-04-17
Created from EPrint's datestamp field
Updated
2021-11-16
Created from EPrint's last_modified field

Caltech Custom Metadata

Caltech groups
LIGO
Other Numbering System Name
LIGO Document
Other Numbering System Identifier
P1800352