Characterization of Gravitational Waves Signals Using Neural Networks

Caramete, A.; Constantinescu, A. I.; Caramete, L. I.; Popescu, T.; Balasov, R. A.; Felea, D.; Rusu, M. V.; Stefanescu, P.; Tintareanu, O. M.

doi:10.48550/arxiv.2009.06109

Cited by 1 publication

(1 citation statement)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Convolutional neural networks have been shown to be capable of identifying gravitational waves and their parameters from binary black holes and binary neutron stars, with performance approaching the matched filtering search currently used by LIGO . In addition, these machine learning (ML) method can also be applied to glitches and noise transients identification [53,[71][72][73][74][75][76], signal classification and parameter estimation [77][78][79][80][81], data denoising [82,83], etc. While these works exhibit neural networks that could approach the performance of matched filtering, they are still often applied as or considered "black box" models.…”

Section: Introductionmentioning

confidence: 99%

Generalized Approach to Matched Filtering using Neural Networks

Yan,

Avagyan,

Colgan

et al. 2021

Preprint

View full text Add to dashboard Cite

Gravitational wave science is a pioneering field with rapidly evolving data analysis methodology currently assimilating and inventing deep learning techniques. The bulk of the sophisticated flagship searches of the field rely on the time-tested matched filtering principle within their core. In this paper, we make a key observation on the relationship between the emerging deep learning and the traditional techniques: matched filtering is formally equivalent to a particular neural network. This means that a neural network can be constructed analytically to exactly implement matched filtering, and can be further trained on data or boosted with additional complexity for improved performance. This fundamental equivalence allows us to define a "complexity standard candle" allowing us to characterize the relative complexity of the different approaches to gravitational wave signals in a common framework. Additionally it also provides a glimpse of an intriguing symmetry that could provide clues on how neural networks approach the problem of finding signals in overwhelming noise. Moreover, we show that the proposed neural network architecture can outperform matched filtering, both with or without knowledge of a prior on the parameter distribution. When a prior is given, the proposed neural network can approach the statistically optimal performance. We also propose and investigate two different neural network architectures MNet-Shallow and MNet-Deep, both of which implement matched filtering at initialization and can be trained on data. MNet-Shallow has simpler structure, while MNet-Deep is more flexible and can deal with a wider range of distributions. Our theoretical findings are corroborated by experiments using real LIGO data and synthetic injections. Finally, our results suggest new perspectives on the role of deep learning in gravitational wave detection.

show abstract