TY - JOUR

T1 - Novel neural-network architecture for continuous gravitational waves

AU - Joshi, Prasanna M.

AU - Prix, Reinhard

PY - 2023/9/18

Y1 - 2023/9/18

N2 - The high computational cost of wide-parameter-space searches for continuous gravitational waves (CWs) significantly limits the achievable sensitivity. This challenge has motivated the exploration of alternative search methods, such as deep neural networks (DNNs). Previous attempts [1,2] to apply convolutional image-classification DNN architectures to all-sky and directed CW searches showed promise for short, one-day search durations, but proved ineffective for longer durations of around ten days. In this paper, we offer a hypothesis for this limitation and propose new design principles to overcome it. As a proof of concept, we show that our novel convolutional DNN architecture attains matched-filtering sensitivity for a targeted search (i.e., single sky-position and frequency) in Gaussian data from two detectors spanning ten days. We illustrate this performance for two different sky positions and five frequencies in the 20-1000 Hz range, spanning the spectrum from an "easy"to the "hardest"case. The corresponding sensitivity depths fall in the range of 82-86/ Hz. The same DNN architecture is trained for each case, taking between 4-32 hours to reach matched-filtering sensitivity. The detection probability of the trained DNNs as a function of signal amplitude varies consistently with that of matched filtering. Furthermore, the DNN statistic distributions can be approximately mapped to those of the F-statistic under a simple monotonic function.

AB - The high computational cost of wide-parameter-space searches for continuous gravitational waves (CWs) significantly limits the achievable sensitivity. This challenge has motivated the exploration of alternative search methods, such as deep neural networks (DNNs). Previous attempts [1,2] to apply convolutional image-classification DNN architectures to all-sky and directed CW searches showed promise for short, one-day search durations, but proved ineffective for longer durations of around ten days. In this paper, we offer a hypothesis for this limitation and propose new design principles to overcome it. As a proof of concept, we show that our novel convolutional DNN architecture attains matched-filtering sensitivity for a targeted search (i.e., single sky-position and frequency) in Gaussian data from two detectors spanning ten days. We illustrate this performance for two different sky positions and five frequencies in the 20-1000 Hz range, spanning the spectrum from an "easy"to the "hardest"case. The corresponding sensitivity depths fall in the range of 82-86/ Hz. The same DNN architecture is trained for each case, taking between 4-32 hours to reach matched-filtering sensitivity. The detection probability of the trained DNNs as a function of signal amplitude varies consistently with that of matched filtering. Furthermore, the DNN statistic distributions can be approximately mapped to those of the F-statistic under a simple monotonic function.

UR - http://www.scopus.com/inward/record.url?scp=85172785363&partnerID=8YFLogxK

U2 - 10.48550/arXiv.2305.01057

DO - 10.48550/arXiv.2305.01057

M3 - Article

AN - SCOPUS:85172785363

VL - 108

JO - Physical Review D

JF - Physical Review D

SN - 2470-0010

IS - 6

M1 - 063021

ER -