English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

The ability of deep learning (DL) approaches to learn generalised signal andnoise models, coupled with their fast inference on GPUs, holds great promisefor enhancing gravitational-wave (GW) searches in terms of speed, parameterspace coverage, and search sensitivity. However, the opaque nature of DL modelsseverely harms their reliability. In this work, we meticulously develop a DLmodel stage-wise and work towards improving its robustness and reliability.First, we address the problems in maintaining the purity of training data byderiving a new metric that better reflects the visual strength of the 'chirp'signal features in the data. Using a reduced, smooth representation obtainedthrough a variational auto-encoder (VAE), we build a classifier to search forcompact binary coalescence (CBC) signals. Our tests on real LIGO data show animpressive performance of the model. However, upon probing the robustness ofthe model through adversarial attacks, its simple failure modes wereidentified, underlining how such models can still be highly fragile. As a firststep towards bringing robustness, we retrain the model in a novel frameworkinvolving a generative adversarial network (GAN). Over the course of training,the model learns to eliminate the primary modes of failure identified by theadversaries. Although absolute robustness is practically impossible to achieve,we demonstrate some fundamental improvements earned through such training, likesparseness and reduced degeneracy in the extracted features at different layersinside the model. We show that these gains are achieved at practically zeroloss in terms of model performance on real LIGO data before and after GANtraining. Through a direct search on 8.8 days of LIGO data, we recover twosignificant CBC events from GWTC-2.1, GW190519_153544 and GW190521_074359. Wealso report the search sensitivity obtained from an injection study.

Towards a robust and reliable deep learning approach for detection of compact binary mergers in gravitational wave data