Employing deep learning for detection of gravitational waves from compact binary coalescences

Abstract

The matched filtering paradigm is the mainstay of gravitational wave (GW) searches from astrophysical coalescing compact binaries (e.g., binary black holes, binary neutron star, neutron star, and black hole pair). The compact binary coalescence (CBC) search pipelines (e.g. GstLAL, PyCBC) perform the matched filter between the GW detector’s data and a large set of analytical waveforms. However, the computational cost of performing matched filter is very high as the required number of the analytical waveforms is also high. Recently, various deep learning-based methods have been deployed to identify a GW signal in the output of the detector in contrast to matched filtering technique which is computationally quite expensive. In past work, the researchers have considered the detection of GW signal mainly as a classification problem, in which they train the deep learning-based architecture by considering the noise and the GW signal as two different classes. However, in this work, for the first time, we have combined the Convolutional Neural Network (CNN) and matched filter methods to reduce the computational cost of the search by reducing the number of matched filtering operations. We have implemented the CNN based architecture not only for classification of the signal but also to identify the location of the signal in the intrinsic parameter (e.g., mass, spin) space. Identifying the location (patch) in which the detected signal lies enables us to perform the matched filter operations between the data and the analytical waveforms generated for the smaller region of the parameter space (patch) only - thereby reducing the computational cost of the search. We demonstrate our method for two-dimensional parameter space (i.e., for non-spinning case, in component mass space (m1, m2) only) for stellar to high mass binary black hole (BBH) systems. In particular, we are able to classify between pure noise and noisy BBH signals with 99% accuracy. Further, the detected signals have been sub-classified into patches in mass components with an average accuracy ≥ 97%.