Machine Learning Applications for Physical Sciences
Leading research into the growing Machine Learning arena
The Machine Learning Applications for Physical Sciences (MAPS) research cluster focus on the application of state-of-the-art Machine Learning algorithms for efficient processing, accurate characterisation and robust prediction of signals arising in physical sciences.
Machine learning methods powered by the increasing computing resources enable scientists to study signals from large amounts of noise-dominated data produced by complex physical processes.
The formation of the Data-Intensive Astronomy (DIA) group in ICRAR has, by coincidence, drawn together a number of computer/data scientists, astronomers, engineers and industry leaders with years of experience in this field.
Whilst tackling the data challenges from the Square Kilometre Array (SKA), we have developed machine learning-based approaches with wider applicability. Through contacts across the Engineering and Mathematical Sciences faculty, we have been working with a number of research groups to apply these approaches to address problems involving signal processing, detection and prediction.
- Discovery of Rare, Complex Celestial Sources
- Metocean Wave Prediction for Offshore Structures
- Detection of Gravitational Waves in LIGO Signals
- Prediction of LIGO Background Signals
- Classification of Dynamical States from Observations
- Mitigation and Excision of RFI for Radio Telescopes and LIDAR
Research opportunities are available for prospective undergraduate, Masters and Doctor of Philosophy students. To submit an expression of interest for a research opportunity, fill out our form or email us.
- International Centre for Radio Astronomy Research (ICRAR)
- Faculty of Engineering and Mathematical Sciences
School of Engineering:
School of Physics, Mathematics and Computing:
- Prof Linqing Wen
- Dr Joel Bosveld
- Dr Qi Chu
- W/Prof Mohammed Bennamoun
- Dr Hossein Rahmani
- Dr Senjian An
- W/Prof Michael Small
Centre for Offshore Foundation Systems (COFS):
- Other Members