People
Faculty
Dr. Devon Jarvis
Devon is an Associate Lecturer in the School of Computer Science and Applied Mathematics at Wits. His research focuses on computational neuroscience and theoretical machine learning. His main research directions focus on semantic reasoning and decision making, and computational models of learning and cognitive diseases.
Dr. Geraud Nangue Tasse
Geraud is an Associate Lecturer in the School of Computer Science and Applied Mathematics at Wits. His research focuses on compositional generalisation and safe reinforcement learning.
Dr. Victoria Williams
Victoria is a Postdoctoral Researcher in Computer Science and Applied Mathematics at Wits. Her research integrates neuroscience and artificial intelligence, focusing on how insights from the anatomy, behaviour and evolution of human and animal brains can advance adaptive and intelligent systems.
Prof. Stefano Sarao Mannelli
Stefano is a tunure-track Assistant Professor in the Data Science and AI division in the Computer Science department of Chalmers University of Technology and Gothenburg University, and Visiting Lecturer at the University of the Witwatersrand. His research focuses on analysing machine learning problems using a model-based approach, where the complexity of the problem is reduced to obtain a parsimonious solvable model that still captures the phenomenon of interest. In his previous works, he applied several variations of this approach to study problems in learning, such as transfer learning, continual learning, and curriculum learning.
Masters Students
Nourhan Abdelrhim
I am a Master’s student specializing in medical image classification. My research focuses on leveraging self-supervised learning, curriculum learning, and explainable AI to enhance model accuracy and interpretability.
Christine Bau
My research in the lab focuses on exploring how machine learning can emulate aspects of human communication and understanding. I am particularly interested in modeling the dynamics of conversation between multimodal and unimodal agents, investigating how shared representations can support the emergence of mutual understanding. My work lies mostly in the domain of machine learning, computer vision, and natural language processing.
Liam Culligan
My work in the lab focuses on developing a computational model that integrates the multimodal representational architecture of the Hub-and-Spoke framework with selective episodic memory consolidation mechanisms of the Go-CLS framework. This model aims to simulate how the brain transforms fragmented sensory experiences into coherent, generalisable conceptual knowledge. More broadly, my research interests focus on semantic cognition, memory consolidation, and biologically inspired artificial intelligence, with a particular focus on how structured conceptual representations develop from real-world, multimodal experiences.
Sergio Frasco
I am currently studying the role vision can play in Reinforcement Learning. By decoupling the value function into successor representation and reward maps predicted by vision - I hope to create a reinforcement learning agent that learns more efficiently than what we would see from standard RL agents.
Muhammad Sahal Goolam
Muhammad is a Master's student in Computer Science at Wits. His research focuses on model-based and reinforcement learning, representation learning, and biologically inspired approaches to vision-driven intelligence.
Sibongakonke Kubheka
My research focuses on computational neuroscience and machine learning, specifically within the domain of continual learning, bridging insights from neuroscience to address fundamental challenges in machine learning.
Samuel Winbono-Mpaaba Mba
Samuel Mba is an MSc student in Computer Science at the University of the Witwatersrand, in the School of Computer Science and Applied Mathematics. His research focuses on computational neuroscience and reinforcement learning, with a particular interest in applying Spiking Neural Networks (SNNs) to RL tasks. He is also broadly interested in computer vision and its integration with biologically inspired models.
Raees Moosa
I am a Masters student in the School of Computer Science and Applied Mathematics at Wits. My research extends the VAE architecture with a focus on interpretability and structure of the latent space.
Yonatan Oudmayer
In our lab, I study the mathematical dynamics of convolutional neural networks by extending the gated deep linear framework to capture true non-linear behavior. By combining singular value decomposition with Fourier-based pathway analysis, I explore how data structure and gating operations shed light on non-linear mechanisms in CNNs.More broadly, I’m interested in researching mathematical descriptions of non-linearity in neural networks, with the goal of deepening theoretical insight into their interpretability.