GspyNetTree: distinguishing gravitational-wave signals from detector transient noise for the fourth LIGO-Virgo-KAGRA observing run.

Following the first detection of gravitational-waves (GWs), the LIGO and Virgo collaborations started a new field of astronomy by providing a new way of understanding the Universe. Despite achieving detector sensitivities capable of detecting the extremely small amplitude of GWs, LIGO and Virgo detector data contain frequent bursts of non-Gaussian transient noise, commonly known as 'glitches'. Glitches come in various time-frequency morphologies, and they are particularly challenging when they mimic the form of real GWs.

We developed a new signal-vs-glitch classifier method that leverages a Convolutional Neural Network (CNN) image classifier, Gravity Spy, that successfully classified LIGO and Virgo glitches in previous observing runs. In order to automate validation of LIGO-Virgo GW event candidates, a signal-vs-glitch classifier must be robust to a broad array of background noise, new sources of glitches, and the likely occurrence of overlapping glitches and GWs. In response to these new challenges, we present GSpyNetTree, the Gravity Spy Convolutional Neural Network Decision Tree: a multi-CNN classifier using CNNs in a decision tree sorted via total GW candidate mass.