Robotics, Autonomous Intelligence and Learning Lab
The RAIL Lab, established in 2014, is dedicated to conducting cutting-edge research in the field of artificially intelligent systems. With a focus on both fundamental and applied research, our vision is to serve as a prominent centre of excellence and a hub for AI activities in Africa. We aim to make significant contributions to the field of AI while also applying our findings to benefit society at large.
Latest Research
The Sterkfontein Caves Dataset: A Novel View Synthesis Challenge from the Cradle of Humankind
We introduce the challenging Sterkfontein Caves dataset comprising ten underground scenes from a UNESCO World Heritage Site, and use it …
PyCRM: A Python library for reward machine-based reinforcement learning
Reinforcement Learning (RL) research often models environments as Markov decision processes. Yet many real-world tasks are …
Drowning in Degrees of Freedom: One Agent for Every Task
We describe a research programme aimed at the construction of a single, generally intelligent agent—one competent across all …
Redistribution-based Cost Inference Improves Sparse Safe Offline RL
Safe offline RL typically assumes access to dense per-step cost annotations, but in practice supervisors provide only trajectory-level …
The Goal-Directed Frame for General Agents
Reinforcement learning is often framed around episodic, discounted, or average scalar rewards. While useful, these views miss a core …
An Unreasonably Simple Approach to Safe RL
An important problem in reinforcement learning is designing agents that learn to solve tasks safely in an environment. A common …