Detecting and Denoising Gravitational Waves from Neutron Stars using Deep Learning

We present the second part of our Deep Learning based denoising scheme to detect and denoise astrophysical gravitational waves. Here we extend our analysis to include gravitational waves from coalescing Binary Neutron Stars (BNS) and Neutron Star- Black Hole (NSBH) binaries in addition to the Binary Black Holes (BBH) presented previously. We employ similar strategy to perform the denoising on the detector data, with slightly different Neural-Net architecture. We use a convolutional neural network, designed in the auto-encoder configuration that can detect and denoise astrophysical gravitational waves from compact binary coalescence, orders of magnitude faster than the conventional matched-filtering based detection that is currently employed at advanced LIGO (aLIGO) and LVK collaboration at large. The Neural-Net architecture is such that it learns from the sparse representation of data in the time-frequency domain and constructs a non-linear mapping function that maps this representation into two separate masks for signal and noise, facilitating the separation of the two, from raw data. This approach is the first of its kind to apply machine learning based gravitational wave detection/denoising in the 2D representation of gravitational wave data. We applied our formalism to the first gravitational wave from binary neutron star, GW170817, successfully recovering the signal at all three phases of coalescence at both detectors with very low False Alarm Probability (FAP) . The recovered signal of this very high SNR detection differs slightly from the template waveform due of the presense of strong precession signature present in the denoised waveform. This method can also interpolate and extrapolate between modeled templates and explore gravitational waves that are unmodeled and hence not present in the template bank of signals used in the matched-filtering detection pipelines. Faster and efficient detection schemes, such as this method, will be instrumental as ground based detectors reach their design sensitivity, likely to result in several hundreds of potential detections in a few months of observing runs.