Higher Degree by Research Application Portal

With the commissioning of the Square Kilometre Array, radio observations and particularly dynamic objects (fast radio bursts, supernovae, gamma-ray bursts, etc.), due to the greater capability of the next-generation instruments, are receiving more focus. Simultaneously, the volume of data produced by such instruments makes detecting transients a data-driven endeavour. Transient events often represent some of the universe’s most extreme conditions. Therefore, they are of great scientific interest. Transients are detected early in radio astronomy processing pipelines, with real-time detection as the ultimate goal. Spiking Neural Networks (SNNs) borrow more heavily from biological inspiration than Artificial Neural Networks (ANNs); most importantly, their dynamics vary in time, making them helpful in processing spatiotemporal data, like radio telescope visibility data.

Spiking Neural Networks are a very new technology for radio astronomy, with few published results to date. This project would build on recent work tasking SNNs with detecting Radio Frequency Interference (RFI). Transient detection events occur at a similar stage in the processing pipeline and often present similarly to RFI. When paired with neuromorphic computing hardware that efficiently executes SNNs, massive energy- efficiency gains could be realised, translating into considerable operational benefits in radio astronomy. This project is, therefore, significant from an engineering and scientific perspective. This PhD would involve developing and testing SNN architectures to detect transient events in radio astronomy visibility data, applied to simulated and real data from existing instruments.

Please contact Andreas Wicenec in the first instance.

The ideal candidate will hold a Bachelor Honours degree or Masters degree in Astronomy with some experience in compu- tational methods; or a Bachelor Honours degree or Masters degree in Computer Science with an interest in Astronomy and High Performance Computing.

Skills

• Python

• Pytorch, Tensorflow or other Deep Learning experiences favoured

• High Performance computing environments (Slurm, MPI) also appreciated 

Training and Development

The candidate will be responsible for designing and evaluating SNN architectures, learning how to perform large-scale SNN simulation and deep learning experiments on HPC platforms, and potentially collaborating with SynSense, a provider of commercial neuromorphic hardware. 

The Doctor of Philosophy (PhD) is a program of independent, supervised research that is assessed solely on the basis of a thesis, sometimes including a creative work component, that is examined externally. The work presented for a PhD must be a substantial and original contribution to scholarship, demonstrating mastery of the subject of interest as well as an advance in that field of knowledge. 

Visit the course webpage for full details of this course including admission requirements, course rules and the relevant CRICOS code/s.