Benjamin van Niekerk
Papers
1
Total Citations
10
H-Index
1
About
Benjamin van Niekerk is a researcher at the forefront of reinforcement learning (RL), with a focused expertise in developing safe, efficient algorithms for robotic systems. His work primarily addresses the critical challenge of applying RL to continuous state and action spaces under real-world constraints, such as limited time and computational budgets. His most-cited paper, "Online Constrained Model-based Reinforcement Learning" (2020, 10 citations), introduces a framework that enables robots to make robust decisions while adhering to safety and resource limitations—a key step toward deploying RL in physical environments. Beyond this, van Niekerk’s contributions emphasize the integration of model-based approaches with online learning, allowing systems to adapt dynamically without violating operational boundaries. His research is particularly notable for bridging theoretical advances in constrained optimization with practical robotic applications, making it invaluable for students and engineers working on autonomous systems. By tackling the dual demands of performance and safety, van Niekerk is shaping a future where RL agents can operate reliably in the real world.
Research Focus
Key Achievements
Top Papers
- 1Online Constrained Model-based Reinforcement Learning10 citations · 2020