CaltechAUTHORS
  A Caltech Library Service

Application of machine learning algorithms to the study of noise artifacts in gravitational-wave data

Biswas, Rahul and Blackburn, Lindy and Cao, Junwei and Essick, Reed and Hodge, Kari Alison and Katsavounidis, Erotokritos and Kim, Kyungmin and Kim, Young-Min and Le Bigot, Eric-Olivier and Lee, Chang-Hwan and Oh, John J. and Oh, Sang Hoon and Son, Edwin J. and Tao, Ye and Vaulin, Ruslan and Wang, Xiaoge (2013) Application of machine learning algorithms to the study of noise artifacts in gravitational-wave data. Physical Review D, 88 (6). Art. No. 062003. ISSN 1550-7998. http://resolver.caltech.edu/CaltechAUTHORS:20131106-110559505

[img]
Preview
PDF - Published Version
See Usage Policy.

2584Kb
[img]
Preview
PDF - Submitted Version
See Usage Policy.

2226Kb

Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:20131106-110559505

Abstract

The sensitivity of searches for astrophysical transients in data from the Laser Interferometer Gravitational-wave Observatory (LIGO) is generally limited by the presence of transient, non-Gaussian noise artifacts, which occur at a high enough rate such that accidental coincidence across multiple detectors is non-negligible. These “glitches” can easily be mistaken for transient gravitational-wave signals, and their robust identification and removal will help any search for astrophysical gravitational waves. We apply machine-learning algorithms (MLAs) to the problem, using data from auxiliary channels within the LIGO detectors that monitor degrees of freedom unaffected by astrophysical signals. Noise sources may produce artifacts in these auxiliary channels as well as the gravitational-wave channel. The number of auxiliary-channel parameters describing these disturbances may also be extremely large; high dimensionality is an area where MLAs are particularly well suited. We demonstrate the feasibility and applicability of three different MLAs: artificial neural networks, support vector machines, and random forests. These classifiers identify and remove a substantial fraction of the glitches present in two different data sets: four weeks of LIGO’s fourth science run and one week of LIGO’s sixth science run. We observe that all three algorithms agree on which events are glitches to within 10% for the sixth-science-run data, and support this by showing that the different optimization criteria used by each classifier generate the same decision surface, based on a likelihood-ratio statistic. Furthermore, we find that all classifiers obtain similar performance to the benchmark algorithm, the ordered veto list, which is optimized to detect pairwise correlations between transients in LIGO auxiliary channels and glitches in the gravitational-wave data. This suggests that most of the useful information currently extracted from the auxiliary channels is already described by this model. Future performance gains are thus likely to involve additional sources of information, rather than improvements in the classification algorithms themselves. We discuss several plausible sources of such new information as well as the ways of propagating it through the classifiers into gravitational-wave searches.


Item Type:Article
Related URLs:
URLURL TypeDescription
http://dx.doi.org/10.1103/PhysRevD.88.062003 DOIArticle
http://link.aps.org/doi/10.1103/PhysRevD.88.062003PublisherArticle
http://arxiv.org/abs/1303.6984arXivDiscussion Paper
Additional Information:© 2013 American Physical Society. Received 29 April 2013; published 23 September 2013. L.B. was supported by an appointment to the NASA Postdoctoral Program at Goddard Space Flight Center, administered by Oak Ridge Associated Universities through a contract with NASA. K.K., Y.M.K., C. H.L., J. J.O., S.H. O., and E. J. S. were supported in part by the Global Science experimental Data hub Center (GSDC) at KISTI. K. K., Y.M.K., and C.H.L. were supported in part by National Research Foundation Grant funded by the Korean Government (NRF-2011-220-C00029). C. H. L. was supported in part by the BAERI Nuclear R&D program (M20808740002). J.C., E.O. L., and X.W. were supported in part by the Ministry of Science and Technology of China under the National 973 Basic Research Program (Grants No. 2011CB302505 and No. 2011CB302805). T.Y. was supported in part by the National High-Tech Research and Development Plan of China (Grant No. 2010AA012302). R. E., K.H., E. K., and R.V. were supported by LIGO laboratory. LIGO was constructed by the California Institute of Technology and Massachusetts Institute of Technology with funding from the National Science Foundation and operates under cooperative agreement PHY-0757058.
Group:LIGO
Funders:
Funding AgencyGrant Number
NASA Postdoctoral ProgramUNSPECIFIED
KISTI Global Science experimental Data hub Center (GSDC) UNSPECIFIED
National Research Foundation of KoreaNRF-2011-220-C00029
BAERI Nuclear R&D programM20808740002
Ministry of Science and Technology of China National 973 Basic Research Program2011CB302505
Ministry of Science and Technology of China National 973 Basic Research Program2011CB302805
National High-Tech Research and Development Plan of China2010AA012302
LIGO LaboratoryUNSPECIFIED
NSFPHY-0757058
Classification Code:PACS: 04.80.Nn, 07.05.Mh, 07.05.Kf
Record Number:CaltechAUTHORS:20131106-110559505
Persistent URL:http://resolver.caltech.edu/CaltechAUTHORS:20131106-110559505
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:42270
Collection:CaltechAUTHORS
Deposited By: Jason Perez
Deposited On:06 Nov 2013 23:59
Last Modified:21 Jun 2017 23:31

Repository Staff Only: item control page