Research

My research interests ㅣie at the interface of optimization, statistics, and machine learning, with a focus on developing theories and efficient algorithms for data-assisted decision making. I am especially interested in uncovering the geometric principles that govern low-complexity models in high-dimensional settings and exploring how these insights can explain the outstanding performance of modern machine learning and AI technologies, and advance decision-making tasks. Currently, my efforts focus on

• Geometry of low-complexity models for high-dimensional statistics and learning
  • Identifying structural patterns and geometric properties in data-driven models
  • Building efficient computational frameworks for leveraging these structures
  • Establishing theoretical limits and performance guarantees
• Decision making with high-dimensional data
  • Designing predictive strategies with noisy or non-Euclidean data
  • Causal inference with overparameterized models
• Machine learning/Deep learning theory
  • General machine learning theory
  • Science of deep learning: training dynamics, scaling laws, …
  • Any topics related to understanding deep learning models