Higher Degree by Research Application Portal

The imminent construction of next-generation radio observatories, such as the Square Kilometre Array (SKA), aims to revolutionise our understanding of the universe by generating orders of magnitude more observations. The output of an observation is a vast three-dimensional image cube spanning spatial, spectral, and temporal dimensions. Source detection is a core challenge in producing radio-sky surveys based on these image cubes. Traditional algorithms sift through these sparsely populated image cubes and may struggle with the scale and complexity of future datasets. Experimental deep learning approaches perform well but require extraordinary computing resources to operate on a practical scale.

Neuromorphic computing and Spiking Neural Networks (SNNs) could help solve this challenge. SNNs are a very new technology for radio astronomy, with only few published results to date. By emulating the time-varying behaviour of biological neurons, SNNs can handle spatio-temporal information more easily than their artificial counterparts. This project would involve adapting ANN-based source detection techniques to SNNs and inventing novel purely SNN-based methods to determine their potential performance and efficiency benefits. 

This is an opportunity for a PhD in a rapidly growing new area in Machine Learning, using ML model that much more closely mimics the operation of a real brain, by reproducing the spike-triggering nature of real neurons. The PhD will collaborate closely with a local start up that is developing finely tuned hyper-efficient implementations of these networks.

Please contact Andreas Wicenec in the first instance.

The ideal candidate will hold a Bachelor Honours or Masters degree in Astronomy with some experience in compu- tational methods; or a Bachelor Honours 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 environment experience (Slurm, MPI) also appreciated

Training and Development

The candidate will be responsible for designing and evaluating SNN models and learning how to perform large-scale SNN simulations and deep learning experiments on HPC platforms. Possibility of collaboration 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.