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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

GWSkyNet-Multi is a machine learning model developed for classification ofcandidate gravitational-wave events detected by the LIGO and Virgoobservatories. The model uses limited information released in the low-latencyOpen Public Alerts to produce prediction scores indicating whether an event isa merger of two black holes, a merger involving a neutron star, or anon-astrophysical glitch. This facilitates time sensitive decisions aboutwhether to perform electromagnetic follow-up of candidate events duringLIGO-Virgo-KAGRA (LVK) observing runs. However, it is not well understood howthe model is leveraging the limited information available to make itspredictions. As a deep learning neural network, the inner workings of the modelcan be difficult to interpret, impacting our trust in its validity androbustness. We tackle this issue by systematically perturbing the model and itsinputs to explain what underlying features and correlations it has learned fordistinguishing the sources. We show that the localization area of the 2D skymaps and the computed coherence versus incoherence Bayes factors are used asstrong predictors for distinguishing between real events and glitches. Theestimated distance to the source is further used to discriminate between binaryblack hole mergers and mergers involving neutron stars. We leverage thesefindings to show that events misclassified by GWSkyNet-Multi in LVK's thirdobserving run have distinct sky area, coherence factor, and distance valuesthat influence the predictions and explain these misclassifications. Theresults help identify the model's limitations and inform potential avenues forfurther optimization.

Explaining the GWSkyNet-Multi machine learning classifier predictions for gravitational-wave events