Hi, I am Dian Jin (金典), a Ph.D. student in Electrical and Computer Engineering at the University of Wisconsin–Madison, advised by Prof. Jeremy Coulson.

My research lies at the intersection of machine learning, optimization, and dynamical systems. I develop data-driven methods for learning predictive representations from trajectory data, quantifying their uncertainty, and using them for efficient robotic control. My work combines theoretical analysis, algorithm development, and computational experiments. I received my B.S. in Mathematics from Soochow University in 2022 and my M.A. in Mathematics from UW–Madison in 2023. Outside research, I enjoy photography and exploring the outdoors.

📖 Research Interests

• Machine Learning for Dynamical Systems
  • Online representation and subspace learning
  • Data-driven prediction and control
  • Uncertainty quantification and robustness
• Optimization and Algorithmic Foundations
  • Online and nonconvex optimization
  • Optimization on Grassmann and flag manifolds
  • Low-rank methods for high-dimensional data
• Robot Learning and Control
  • Learning-based and data-enabled predictive control
  • Learning from offline demonstrations
  • Robotic manipulation and simulation
• Multimodal Foundation Models
  • Parameter-efficient adaptation
  • Low-rank representation learning
  • Vision–language modeling

📝 Recent Publications

• On the sensitivity of the subspace predictor to behavioral perturbations Dian Jin, Jeremy Coulson. IEEE Control Systems Letters.
• Online subspace learning on flag manifolds for system identification Dian Jin, Jeremy Coulson. 8th Annual Learning for Dynamics & Control Conference.
• Parameter-efficient tuning of large-scale multimodal foundation model Haixin Wang, Xinlong Yang, Jianlong Chang, Dian Jin, Jinan Sun, Shikun Zhang, Xiao Luo, Qi Tian. Advances in Neural Information Processing Systems, 36:15752-15774, 2023.

📄 Academic Services

• Reviewer for: L4DC’25, L4DC’26, CDC’26, Automatica