Youngjae Min

Youngjae Min is a graduate student in aeronautics and astronautics whose research focuses on creating rigorous machine learning and control techniques for safety-critical systems. His work tackles the challenge of enabling autonomous systems to operate reliably in uncertain and rapidly changing environments—an essential requirement for applications such as advanced air mobility, smart grids, and robotic systems. Youngjae created HardNet, a neural network architecture that guarantees constraint satisfaction while maintaining universal approximation capabilities, offering a powerful foundation for learning-based control with built-in safety assurances. As a MathWorks Fellow, he will design machine learning algorithms with provable performance and embed control certificates directly into training protocols to enforce verifiable safety across time. MATLAB, along with the Control System and Robust Control Toolboxes, is instrumental to his work, supporting modeling, simulation, and formal verification. Youngjae’s research could enable a new class of adaptive, trustworthy autonomous systems, with broad implications for industries that demand both intelligence and safety from their technologies.