Detection and classification of supernova gravitational wave signals: A deep learning approach
M. Chan, I. S. Heng, C. Messenger
Abstract
We demonstrate the application of a convolutional neural network to the gravitational wave signals from core collapse supernovae. Using simulated time series of gravitational wave detectors, we show that, based on the explosion mechanisms, a convolutional neural network can be used to detect and classify the gravitational wave signals buried in noise. For the waveforms used in the training of the convolutional neural network, our results suggest that a network of advanced LIGO, advanced VIRGO and KAGRA (HLVK), or a network of LIGO A+, advanced VIRGO and KAGRA ($\mathrm{H}+\mathrm{L}+\mathrm{VL}$) is likely to detect a magnetorotational core collapse supernova out to the large or even small Magellanic clouds. A neutrino-driven event is likely to be detectable with a network of $\mathrm{H}+\mathrm{L}+\mathrm{VK}$ and HLVK if the distance is within 5 and 3 kpc respectively. By testing the convolutional neural network with waveforms not used for training, we show that the true alarm probability is 66% at 60 kpc for waveforms R3E1AC and R4E1FC_L with a network of $\mathrm{H}+\mathrm{L}+\mathrm{VK}$. For waveforms s20 and SFHx, the true alarm probabilities are 56% for $\mathrm{H}+\mathrm{L}+\mathrm{VK}$ at 7 kpc and 50% for HLVK at 5 kpc respectively. All are at false alarm probability equal to 0.1%.