top of page
Astronomy & Astrophysics (Pulsars & Time Domain Astrophysics)
Postgraduate Researcher, Jodrell Bank Centre for Astrophysics, The University of Manchester
Researcher, Bose.X Center for Astrophysical Research
Use of Deep Learning to Classify Lensed and Unlensed Gravitational Waves
1
Jaswant Jayacumaar , Shashwat Singh , Adarsh Mahor , Janvita
BOSE.X-TRIAC, Sardar Vallabhbhai National Institute of Technology, Surat, India
Jodrell Bank Centre for Astrophysics, Department of Physics and Astronomy, The University of Manchester
Manchester M13 9PL United Kingdom
Sorbonne University, Paris, France
2
3
1,2
1,3 1 1
ABSTRACT
Gravitational waves (GWs) are ripples in the curvature of space-time, produced by energetic processes such as the collision of two compact objects orbiting each other and core-collapse supernovae. These GWs are lensed i.e., bent near massive astrophysical objects, similar to electromagnetic waves. Lensing of GWs produces multiple images at different times and it is hard to classify a lensed GW from an unlensed one. Hence, we have decided to use convolutional neural networks (CNN) to extract the complicated features of lensed GWs and use these features to classify lensed GWs from unlensed ones. Furthermore, these features are vital to study the merger events as they possess physical significance.
Key Words: Gravitational Waves, Gravitational Lensing, Deep Learning, Convolutional Neural Networks
This project is being conducted in collaboration with Bose.X Center for Astrophysical Research - A Division of Bose.X
Advisor: Dr. Kamlesh N Pathak
Professor
Physics Department, Sardar Vallabhbhai National Institute of Technology
Surat, Gujarat, India
bottom of page