A Caltech Library Service

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

Coughlin, S. and Bahaadini, S. and Rohani, N. and Zevin, M. and Patane, O. and Harandi, M. and Jackson, C. and Noroozi, V. and Allen, S. and Areeda, J. and Coughlin, M. and Ruiz, P. and Berry, C. P. L. and Crowston, K. and Katsaggelos, A. K. and Lundgren, A. and Østerlund, C. and Smith, J. R. and Trouille, L. and Kalogera, V. (2019) Classifying the unknown: Discovering novel gravitational-wave detector glitches using similarity learning. Physical Review D, 99 (8). Art. No. 082002. ISSN 2470-0010. doi:10.1103/physrevd.99.082002.

[img] PDF - Published Version
See Usage Policy.

[img] PDF - Submitted Version
See Usage Policy.


Use this Persistent URL to link to this item:


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.

Item Type:Article
Related URLs:
URLURL TypeDescription Paper
Coughlin, M.0000-0002-8262-2924
Kalogera, V.0000-0001-9236-5469
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.
Funding AgencyGrant Number
David and Ellen Lee Postdoctoral ScholarshipUNSPECIFIED
CIERA Board of Visitors Research ProfessorshipUNSPECIFIED
Other Numbering System:
Other Numbering System NameOther Numbering System ID
LIGO DocumentP1800352
Issue or Number:8
Record Number:CaltechAUTHORS:20190417-084807411
Persistent URL:
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:94738
Deposited By: Tony Diaz
Deposited On:17 Apr 2019 22:00
Last Modified:16 Nov 2021 17:07

Repository Staff Only: item control page