English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

In the last few years, machine learning techniques, in particularconvolutional neural networks, have been investigated as a method to replace orcomplement traditional matched filtering techniques that are used to detect thegravitational-wave signature of merging black holes. However, to date, thesemethods have not yet been successfully applied to the analysis of longstretches of data recorded by the Advanced LIGO and Virgo gravitational-waveobservatories. In this work, we critically examine the use of convolutionalneural networks as a tool to search for merging black holes. We identify thestrengths and limitations of this approach, highlight some common pitfalls intranslating between machine learning and gravitational-wave astronomy, anddiscuss the interdisciplinary challenges. In particular, we explain in detailwhy convolutional neural networks alone cannot be used to claim a statisticallysignificant gravitational-wave detection. However, we demonstrate how they canstill be used to rapidly flag the times of potential signals in the data for amore detailed follow-up. Our convolutional neural network architecture as wellas the proposed performance metrics are better suited for this task than astandard binary classifications scheme. A detailed evaluation of our approachon Advanced LIGO data demonstrates the potential of such systems as triggergenerators. Finally, we sound a note of caution by constructing adversarialexamples, which showcase interesting "failure modes" of our model, where inputswith no visible resemblance to real gravitational-wave signals are identifiedas such by the network with high confidence.

Convolutional neural networks: a magic bullet for gravitational-wave detection?