Maties Machine Learning (MML) is a seminar series and discussion forum with the goal of bringing together people working on machine learning at Stellenbosch University. We meet roughly every second week for short talks on people’s current work, some ML-related topic, or open discussions. The idea is to get to know what others are working on and to strengthen machine learning research at Stellenbosch.
22 October, 13:00-13:50 via MS Teams (link will be emailed)
Mkhuseli Ngxande - Bias remediation in driver drowsiness detection systems using generative adversarial networks
Datasets are crucial when training a deep neural network. When datasets are unrepresentative, trained models are prone to bias because they are unable to generalize to real-world settings. This is particularly problematic for models trained in specific cultural contexts, which may not represent a wide range of races, and thus fail to generalise. This is a particular challenge for driver drowsiness detection, where many publicly available datasets are unrepresentative as they cover only certain ethnicity groups. Traditional augmentation methods are unable to improve a model’s performance when tested on other groups with different facial attributes, and it is often challenging to build new, more representative datasets. We introduce a novel framework that boosts the performance of detection of drowsiness for different ethnicity groups.
Michael Deyzel - Temporal convolutional networks for skeleton-based action recognition.
Skeleton motion sequences generated by motion capture (mocap) technology provide simple sparse data suitable for human action recognition and understanding. The key to a successful classifier lies in an effective representation that can model both spatial and temporal dynamics of human actions with large intra-class variance. We explore temporal convolutions for this problem and consider what benefits they have over more conventional RNN approaches to extract inherent characteristics from the temporal keypoint data.
29 October, 13:00-13:50 via MS Teams (link will be emailed)
Michael McGrath (Stone Three) - Title TBA