معرفی
Holden Lee is an Assistant Professor in the Department of Applied Mathematics and Statistics at Johns Hopkins University, where he joined in 2022. His research focuses on the theoretical foundations of machine learning, probability, and their intersections with theoretical computer science. He explores probabilistic methods in modern machine learning, including deep learning-based generative models and convergence guarantees for sampling algorithms like Markov Chain Monte Carlo. Prior to JHU, he was a postdoc at Duke University and a Simons Fellow at UC Berkeley. He holds a PhD in Mathematics from Princeton University and degrees from MIT and the University of Cambridge.
Education:
- PhD in Mathematics, Princeton University, 2019
- MASt in Pure Mathematics, University of Cambridge, 2014
- BSc in Mathematics, MIT, 2013
Research Interests:
- Machine Learning Theory
- Probabilistic Sampling Methods
- Generative Models
- Statistical Learning Theory
Articles Trends: Lee’s recent publications (2022–2024) focus on advancing sampling algorithms for complex distributions, analyzing generative models, and improving convergence guarantees for methods like MCMC and score-based diffusion. His work bridges theory and practice, with applications in multimodal data, text generation, and dynamical systems.
Awards:
- Simons Fellow at UC Berkeley (2021)
Advising & Grants: Lee has contributed to projects at NeurIPS and collaborates on research in AI efficiency and theoretical guarantees. He teaches courses on probability and applied mathematics at JHU and Duke.
Labs/Teams: His research group focuses on theoretical machine learning and probabilistic methods, with ongoing projects on scalable sampling algorithms and generative model analysis.
