Moka (Mokhwa) Lee

I am a PhD candidate in Applied Mathematics and Statistics with a strong foundation in Machine Learning, Optimization, Numerical Analysis, and Artificial Intelligence. I’m affiliated with AI Institute and IACS, and supervised by professor Yifan Sun (CS) and Joseph Mitchell (AMS). My goal is to bridge the gap between theory and application by developing robust, interpretable, and scalable ML/AI solutions that address complex problems across diverse domains.

My PhD research focuses on developing scalable second-order optimization methods, particularly Quasi-Newton (QN) algorithms. I have designed novel low-rank, multi-secant, and limited-memory variants of QN methods to address the challenges of large-scale machine learning tasks, including classification and neural networks. These methods are implemented in MATLAB and Python programming languages, and aim to enhance convergence speed and stability while maintaining computational efficiency. For more information and the recent publications, please check my Google Scholar.

In addition to theoretical development, I have hands-on experience applying machine learning techniques to real-world problems across finance, engineering, and e-commerce. My coursework and research have covered a wide range of ML/AI topics, including supervised and unsupervised learning. I integrate these methods into practical solutions, combining them with numerical optimization tools such as PyTorch, OR-Tools, SeDuMi and MOSEK, for more robust and effective performance.