​Jandré Marais, currently a Master's student in Mathematical Statistics at SU's Department of Statistics and Actuarial Science, was the winner of the first Data Science Deep Learning Hackathon hosted by the Machine Intelligence Institute of Africa. Participants had to make use of about 5 000 labelled road images and train a model to detect potholes in a new set of test images. Jandré's winning model obtained a test accuracy of 85.5%.

The Data Science Hackathon, sponsored by IBM, the Machine Intelligence Institute of Africa, and Cortex Logic, took place in Johannesburg on 16 September 2017. It was preceded by the Deep Learning Indaba, where some of the topics included 'Mathematics for Deep Learning', 'Convolutional Neural Networks', and 'Recurrent Neural Networks'. The indaba also included a panel discussion on 'African Machine Learning'.

The hackathon problem was based upon the paper entitled 'Detecting potholes using simple image processing techniques and real-world footage', authored by S. Nienaber, M.J. Booysen and R.S. Kroon.

Participants could take part in teams consisting of five team members at the most, and had to present their solutions in a two-minute talk. Prizes were awarded to the top three teams in terms of their model's prediction accuracy. A prize was also awarded to the best presentation. More information regarding the hackathon may be found at http://machineintelligenceafrica.org/activities/hackathon. Jandré shared his winning solution at https://github.com/jandremarais/PotholeDetection.

• Jandré's Master's thesis is entitled 'Deep Neural Networks for Multi-label Image Classification'.
• The Department of Statistics and Actuarial Science, in collaboration with the Department of Computer Science, has been presenting an honours programme with a Data Science focus since 2015. Following this honours degree, students can continue with a Master's degree with Statistical Learning Theory and related Computer Science modules. Many colleagues at the two departments share a keen research interest in the Data Science and Machine Learning fields, with current emphasis on bioinformatics, classification, clustering, data visualisation, feature selection, natural language processing and robotics.