Invited talk as part of the Early Career Spotlight (ECS) program at IJCAI-ECAI 2026, which highlights early-career researchers who have made significant contributions to AI.
An online talk on why AI systems feel intelligent, and why that impression can be misleading, given to the FEDSA (Frontend Engineering and Design South Africa) community.
A talk on applying AI to data from the Cradle of Humankind, given at the Entropy Symposium at the Fak’ugesi African Digital Innovation Festival.
A talk on why embodied, robot-centred approaches may be needed for real progress in artificial intelligence.
A talk on what skills IT teams need to work effectively with AI and ML tools, given at IT Indaba’s inaugural South African edition.
A talk on embodied approaches to artificial intelligence, given at the Deep Learning Indaba.
A public lecture on applying machine learning to data from the Sterkfontein Caves, part of the Cradle of Humankind.
A talk on using AI tools effectively in academic research, given at the International Relations PhD Retreat.
A talk on applying machine learning to data from the Sterkfontein Caves, part of the Cradle of Humankind, given at ICRA@40 Africa.
A talk on reward design and safety in reinforcement learning, given at the “I Can’t Believe It’s Not Better” workshop at the Reinforcement Learning Conference.
A talk on composing skills concurrently and over time to transfer to unseen tasks without further training.
A talk on composing learned skills to solve multiple tasks without retraining.
An introductory talk on reinforcement learning, given online to the IndabaX Uganda community.
An introductory talk on Monte Carlo methods, given at the Deep Learning Indaba.
A talk on autonomously learning object-centric abstractions for high-level planning, given at a CoE-MaSS (Centre of Excellence in Mathematical and Statistical Sciences) seminar.
A panel discussion on current and future directions in machine learning research, given online to the IndabaX Namibia community.
A talk on learning abstract, portable representations that can be reused for high-level planning across tasks.
An invited lab talk on learning abstract, portable representations that can be reused for high-level planning across tasks.
A talk on combining previously learned value functions to solve new tasks without further training.
A talk analysing the behaviour and biases of Monte Carlo tree search.
A talk on the effectiveness of heavy rollout policies in UCT-based game-playing agents.