How do we deal with very large data sets of high resolution images, in particular in the field of radio astronomy? This question encompasses the scope of a joint PhD project between the University of Groningen (The Netherlands), the University of Stellenbosch (South Africa), and ASTRON, which is the Netherlands Institute for Radio Astronomy.

Modern radio telescopes typically consist of ~100 to a few hundred receiving elements, whose signals are pairwise correlated producing tens of thousands correlations for tens of thousands of frequency channels simultaneously. For a system like the Square Kilometre Array (SKA) this produces a data deluge of ~1 TByte/s. This data may be affected by man-made radio frequency interference (RFI), instrumental failures and other effects that make the data unsuitable for scientific analysis. Even with good anomaly detectors, it is very challenging and time consuming for the telescope operators to assess anomalous data by hand to understand the underlying issue. Feature extraction of anomalous behavior and its indexing / classification could potentially be a powerful tool to further automate telescope operations.

The aforementioned raw data will be transformed to science-ready data products such as image cubes (a 3D data cube showing an intensity map of the sky as a function of frequency). These data products are used by astronomers to answer scientific questions. For the SKA, one of the envisaged surveys is aptly called the billion galaxy HI survey. Doing science with the data from this survey will require a high level of automation. Of particular interest will be finding the non-standard sources that are most likely to drive new scientific discoveries. Again, detection of anomalous features and identification of sources with similar anomalous features could play a key role in automation of the scientific analysis of SKA data.

The aim of this project is to develop tools for the analysis of very large datasets of high resolution images. These tools will be focused on the application of radio astronomy in collaboration with ASTRON. Candidates are invited to write and develop their own research project within the scope of the proposed topic(s) listed below, which are of particular interest given the deluge of data produced by current radio telescopes like MeerKAT and LOFAR and the future SKA.

● Querying large database of visual objects in an efficient and scalable way
● Investigating indexing methodologies that are robust to different image perturbations
● Discover new (anomalic) objects

Being part of a cutting-edge research programme, the selected candidate will receive excellent training in the form of hands-on instruction, advanced courses and summer/winter schools, complemented by workshops on generic research and transferable skills as well as teaching training. The selected PhD candidate will be committed to conduct independent and original scientific research, to report on this research in international publications and conference presentations, and to describe the results of the research in a PhD dissertation, to be completed within four years.

The successful candidate should:

● have a keen interest in pursuing fundamental and interdisciplinary research at the interface of artificial intelligence and computer science on the one hand and astronomy on the other.
● have a master’s degree or equivalent in computer science, AI or another relevant field
● have a strong background in data science (deep learning, ConvNets, anomaly detection, …)
● preferably have experience in software system building and integration (software engineering skills, databases, principles of service orientation and cloud computing, parallel and distributed computing)
● and have a positive attitude towards interdisciplinary work.

Fixed-term contract: 48 months.

The project will be 4 years long where the candidate will spend two years at UG and two years at SU. The selected candidate would be required to enrol as a PhD student at both institutions for all four years. Two years at UG will be supported by a PhD Scholarship and two years at SU will be supported by a SU E&E Departmental Scholarship. Over and above there will be a joint double doctorate agreement with the two Universities. At the UG the candidate would become a member of the Information Systems group of the Computer Science Department and will work under the supervision of Dr. George Azzopardi and Prof. Dimka Karastoyanova, and at the SU the candidate would be a PhD student in the Department of Electrical and Electronic Engineering under the supervision of Prof. Stefan Wijnholds and Dr. Trienko Grobler. Below are more details about the two Universities and the conditions.

University of Groningen (UG). Founded in 1614, it enjoys an international reputation as a dynamic and innovative institution of higher education offering high-quality teaching and research. Flexible study programmes and academic career opportunities in a wide variety of disciplines encourage the 36,000 students and researchers alike to develop their own individual talents. As one of the best research universities in Europe, the University of Groningen has joined forces with other top universities and networks worldwide to become a truly global centre of knowledge.

Conditions at UG. The PhD student will be enrolled in the PhD Scholarship Programme of the Faculty of Science and Engineering of the University of Groningen from where s/he will receive a scholarship of € 2,207 per month (gross) for a period of 2 years. Information about the terms and conditions of the PhD Scholarship Programme can be found via: https://www.rug.nl/education/phd-programmes/phd-scholarship-programme/.

The scholarship PhD student will participate in the Graduate School of Science training programme for PhD students and will draw up a personal training and supervision plan together with his or her supervisors from both UG and SU. The UG Graduate School also provides a progress monitoring scheme to ensure an efficient PhD trajectory resulting in a PhD thesis within 4 years. A Career Perspectives curriculum is part of the training, which aims to prepare students for their (academic or non-academic) careers after the PhD trajectory.

Stellenbosch University (SU). It is a public research university that was founded in 1918. SU’s vision is to be the leading research-intensive university on the African continent by pursuing excellence and remaining at the forefront of its chosen focus areas, by gaining national and international standing based on its research outputs, by being relevant to the needs of the community, and by being enterprising, innovative and self-renewing. SU has a wide variety of study programmes and a student corpus of about 30,000 students. It is regarded as one of the top research institutions in Africa.

Conditions at SU. The PhD student will simultaneously be enrolled in the SU Faculty of Engineering, Department of Electrical and Electronic Engineering from where s/he will receive a scholarship of R140 000 (net)per annum for a period of 2 years and will carry out his/her PhD research according to the agreed training and supervision plan. In addition to the UG Graduate School of Science and Engineering training programme, the scholarship PhD student will also be able to participate in any SU Postgraduate Skills Development training opportunities on offer.

Faculty of Science and Engineering

University of Groningen (UG). Founded in 1614, it enjoys an international reputation as a dynamic and innovative institution of higher education offering high-quality teaching and research. Flexible study programmes and academic career opportunities in a wide variety of disciplines encourage the 36,000 students and researchers alike to develop their own individual talents. As one of the best research universities in Europe, the University of Groningen has joined forces with other top universities and networks worldwide to become a truly global centre of knowledge.