Postgraduate Studies in Computer & Electronic Engineering

 

Artificial intelligence:

 

The MuST research group offers a postgraduate programme focussed on machine learning. To qualify, prospective students must have an undergraduate degree in a related Engineering field and an interest in machine learning, deep neural networks, statistical pattern recognition and/or generalisation within the context of Artificial Intelligence.

 

MuST students either study the essence of the learning process of different types of deep networks, or apply and improve these techniques in the context of a specific application domain. In practice, students work with MuST researchers on one of our research projects, using popular deep learning tools (such as Pytorch and Tensorflow) to explore specific questions on new and existing data sets. Our students learn about machine learning algorithms, development of software, and the design and interpretation of machine learning experiments.

 

Through its affiliation with the South African Centre for Artificial Intelligence Research (CAIR), MuST is able to offer a number of M.Eng and PhD bursaries annually. We consider exceptional applications throughout the year (space allowing), and have an open application process in July-September.

 

Examples of current topics:

  • The role of depth in ReLU-activated feedforward neural networks.
  • Activation functions and their effect on the generalisation ability of networks.
  • Node activation in deep convolutional networks.
  • Applying recurrent networks to space weather prediction.

MuST also performs application-oriented research in collaboration with domain experts from industry, to solve specific problems using deep learning approaches. Currently students in this subprogramme work in the domains including natural language processing (NLP), space weather, aerodynamics and intelligent value chains. 

Study leaders: 

Prof Marelie Davel, Prof Etienne Barnard & Dr Stefan Lotz. Domain experts participate as co-supervisors where relevant.

Recent dissertations and theses:
Arnold M. Pretorius, Activation functions in deep neural networks. Potchefstroom: North-West University. (Dissertation - M.Eng.), 2019.