UNIVERSITY OF BUCHAREST
FACULTY OF PHYSICS

Guest
2024-11-23 18:12
 HOME     CONFERENCES     SEARCH            LOGIN     NEW USER     IMAGES   


Conference: Bucharest University Faculty of Physics 2022 Meeting


Section: Atomic and Molecular Physics. Astrophysics. Applications


Title:
Deep Learning Techniques for Gravitational Waves Analysis


Authors:
Vlad-Andrei BÂSCEANU (1,2), Laurențiu-Ioan CARAMETE (1), Ana CARAMETE (1), Maria Cătălina IȘFAN (1,2), Alexandru JIPA (2)


*
Affiliation:
1) Institute of Space Science, Măgurele, Romania

2) Faculty of Physics, University of Bucharest, Măgurele, Romania


E-mail
vabasceanu@spacescience.ro


Keywords:
Time-Domain Astrophysics, Deep Learning, Gravitational Waves


Abstract:
In the past years, Deep Learning (DL) has become more and more popular, partially due to the technological development of deep learning specific GPUs (Graphics Processing Unit) but more importantly due to the broad range of uses such as image processing, medical diagnoses and market forecasting. Recently, Deep Learning Convolutional Neural Networks (DL-CNN) were used to detect Gravitational Wave (GW) signals which proved the potential of these algorithms in the context of time-domain astrophysics. In this work we present the preliminary results using different deep learning techniques used in the analysis of GW signals emitted by Massive Black Hole Binaries (MBHB), with component mass ratios in the range of 1-1501. The data consists of simulated GW signals, both clean and injected into Gaussian Random Noise and also randomly generated Gaussian Noise. These results have direct implications in sending early alerts in the context of multi-messenger astronomy.


References:

1) Carrillo M., Gonzalez J. A., Gracia-Linares M., Guzman F. S., „Time series analysis of Gravitational-Wave signals using neural networks”, Journal of Physiscs Conference Series volume 654, article number 012001, 2015

2) Gebhard Timothy D., Kilbertus Niki, Harry Ian W., Schoelkopf Bernhard, „Convolutional neural networks: A magic bullet for gravitational-wave detection?”, Physical Review D volume 100, article number 063015, 2019