Machine Learning Detection of Gravitational Waves

co-supervisor: Linqing Wen (UWA, Physics)

ICRAR and the Physics Department at UWA are working together to develop innovative deep learning approaches to the identification and classification of gravitational waves (GWs) produced by the binary coalescences of neutron stars and stellar-mass black holes.  More than 90 such events have been detected by the LIGO-Virgo Collaboration from the past three science runs.  The next science run is schedule in May 2024 that will last for 18 months.  More discoveries of GWs are expected.

The PhD will build on existing development at the UWA to improve the detection sensitivity of deep learning methods and to search for sub-threshold gravitational wave events in the archived LVC GW data.  The PhD is expected to work with researchers at ICRAR, Physics Department at UWA, as well as a number of leading AI institutes.

There are possibilities to use the new population studies to investigate: theoretical models of stellar evolution and the origins of Black Holes, or the equations of state of bodies at the extremes of physics.

Machine Learning is a massively dynamic field of research and the PhD candidate will explore and advance the cutting edge developments in ML to find the best match between the problem and the solution. There are opportunities for the candidate to be at the forefront of ML research and engage with our industry partners.

   Successful GW detections and the ML model used in the
recent Physical Review D paper by our team (10.1103/PhysRevD.104.064046)