S. B. Coughlin

Gravity Spy: Integrating Advanced LIGO Detector Characterization, Machine Learning, and Citizen Science

With the first direct detection of gravitational waves, the advanced laser interferometer gravitational-wave observatory (LIGO) has initiated a new field of astronomy by providing an alternative means of sensing the universe. The extreme sensitivity required to make such detections is achieved through exquisite isolation of all sensitive components of LIGO from non-gravitational-wave disturbances. Nonetheless, LIGO is still susceptible to a variety of instrumental and environmental sources of noise that contaminate the data. Of particular concern are noise features known as glitches, which are transient and non-Gaussian in their nature, and occur at a high enough rate so that accidental coincidence between the two LIGO detectors is non-negligible. Glitches come in a wide range of time-frequency-amplitude morphologies, with new morphologies appearing as the detector evolves. Since they can obscure or mimic true gravitational-wave signals, a robust characterization of glitches is paramount in the effort to achieve the gravitational-wave detection rates that are predicted by the design sensitivity of LIGO. This proves a daunting task for members of the LIGO Scientific Collaboration alone due to the sheer amount of data. In this paper we describe an innovative project that combines crowdsourcing with machine learning to aid in the challenging task of categorizing all of the glitches recorded by the LIGO detectors. Through the Zooniverse platform, we engage and recruit volunteers from the public to categorize images of time-frequency representations of glitches into pre-identified morphological classes and to discover new classes that appear as the detectors evolve. In addition, machine learning algorithms are used to categorize images after being trained on human-classified examples of the morphological classes. Leveraging the strengths of both classification methods, we create a combined method with the aim of improving the efficiency and accuracy of each individual classifier. The resulting classification and characterization should help LIGO scientists to identify causes of glitches and subsequently eliminate them from the data or the detector entirely, thereby improving the rate and accuracy of gravitational-wave observations. We demonstrate these methods using a small subset of data from LIGOs first observing run.

ASTROPHYSICAL PRIOR INFORMATION AND GRAVITATIONAL-WAVE PARAMETER ESTIMATION

The detection of electromagnetic counterparts to gravitational waves (GWs) has great promise for the investigation of many scientific questions. While it is well known that certain orientation parameters can reduce uncertainty in other related parameters, it was also hoped that the detection of an electromagnetic signal in conjunction with a GW could augment the measurement precision of the mass and spin from the gravitational signal itself. That is, knowledge of the sky location, inclination, and redshift of a binary could break degeneracies between these extrinsic, coordinate-dependent parameters and the physical parameters that are intrinsic to the binary. In this paper, we investigate this issue by assuming perfect knowledge of extrinsic parameters, and assessing the maximal impact of this knowledge on our ability to extract intrinsic parameters. We recover similar gains in extrinsic recovery to earlier work; however, we find only modest improvements in a few intrinsic parameters—namely the primary components spin. We thus conclude that, even in the best case, the use of additional information from electromagnetic observations does not improve the measurement of the intrinsic parameters significantly.

Constraints on the neutron star equation of state from AT2017gfo using radiative transfer simulations

The detection of the binary neutron star merger GW170817 together with the observation of electromagnetic counterparts across the entire spectrum inaugurated a new era of multimessenger astronomy. In this study, we incorporate wavelength-dependent opacities and emissivities calculated from atomic-structure data enabling us to model both the measured light curves and spectra of the electromagnetic transient AT2017gfo. Best fits of the observational data are obtained by Gaussian Process Regression, which allows us to present posterior samples for the kilonova and source properties connected to GW170817. Incorporating constraints obtained from the gravitational wave signal measured by the LIGO-Virgo Scientific Collaboration, we present a 90 per cent upper bound on the mass ratio q ≲ 1.38 and a lower bound on the tidal deformability of Λ ≳ 197, which rules out sufficiently soft equations of state. Our analysis is a path-finder for more realistic kilonova models and shows how the combination of gravitational wave and electromagnetic measurements allow for stringent constraints on the source parameters and the supranuclear equation of state.

Improvements in Gravitational-Wave Sky Localization with Expanded Networks of Interferometers

A milestone of multi-messenger astronomy has been achieved with the detection of gravitational waves from a binary neutron star merger accompanied by observations of several associated electromagnetic counterparts. Joint observations can reveal details of the engines that drive the electromagnetic and gravitational-wave emission. However, locating and identify an electromagnetic counterparts to a gravitational-wave event is heavily reliant on localization of the source through gravitational-wave information. We explore the sky localization of a simulated set of neutron star mergers as the worldwide network of gravitational-wave detectors evolves through the next decade, performing the first such study for neutron star -- black hole binary sources. Currently, three detectors are observing with additional detectors in Japan and India expected to become operational in the coming years. With three detectors, we recover a median neutron star -- black hole binary sky localization of 60 deg

Deep multi-view models for glitch classification

^2

Exploring a search for long-duration transient gravitational waves associated with magnetar bursts

at the 90\% credible level. As all five detectors become operational, sources can be localized to a median of 11 deg

Direct: Deep Discriminative Embedding for Clustering of Ligo Data.

^2

Machine learning for Gravity Spy: Glitch classification and dataset

on the sky.

Characterizing accreting double white dwarf binaries with LISA and Gaia

Non-cosmic, non-Gaussian disturbances known as “glitches”, show up in gravitational-wave data of the Advanced Laser Interferometer Gravitational-wave Observatory, or aLIGO. In this paper, we propose a deep multi-view convolutional neural network to classify glitches automatically. The primary purpose of classifying glitches is to understand their characteristics and origin, which facilitates their removal from the data or from the detector entirely. We visualize glitches as spectrograms and leverage the state-of-the-art image classification techniques in our model. The suggested classifier is a multi-view deep neural network that exploits four different views for classification. The experimental results demonstrate that the proposed model improves the overall accuracy of the classification compared to traditional single view algorithms.

Distinguishing neutron stars from black holes and probing the mass gap with Advanced LIGO/Virgo observations

Soft gamma repeaters and anomalous X-ray pulsars are thought to be magnetars, neutron stars with strong magnetic fields of order