Background reduction in LIGO-Virgo searches for supernova signals with machine learning

About 20% of the data collected by Advanced LIGO and Virgo in the next observing runs will be single-interferometer data, i.e., they will be collected at times when only one detector in the network is operating in observing mode. If a galactic supernova occurs during single-interferometer times, separation of its gravitational-wave signal from noise will be even more difficult due to lack of coherence between detectors. We present a method to improve the background of LIGO and Virgo single-interferometer supernova searches based on the standard LIGO-Virgo coherent WaveBurst (cWB) pipeline and genetic programming, a supervised machine learning algorithm that uses the strategy of natural selection to solve complex problems. We show that it is possible to discriminate galactic gravitational-wave supernova signals from noise transients with high efficiency, thus increasing the supernova detection reach of Advanced LIGO and Virgo.