Liberal Arts and Sciences
Duan Lab
Led by Leo Li Duan · Statistics
What we study
The Duan Lab develops statistical and machine learning tools for efficient computing and uncertainty quantification, combining Bayesian and optimization methods. Their work applies to neuroscience, engineering, forensics, and data privacy.
What working here is like
- Computational
- No organisms (data only)
- Coding required
- Papers
- 5
Recent papers, explained
- Statistical Modeling of Combinatorial Response Data
2026 · Journal of the American Statistical Association
Proposes a new method to model combinatorial response data, such as survey responses with skip logic.
- Exact sampling of spanning trees via fast-forwarded random walks
2025 · Biometrika
Introduces a new algorithm to sample spanning trees more efficiently, avoiding slow mixing in random walks.
Show 3 more papersShow fewer
- Gibbs Sampling Using Anti-Correlation Gaussian Data Augmentation, with Applications to L1-Ball-Type Models
2025 · Journal of Computational and Graphical Statistics
Develops a faster Gibbs sampling method for models with L1-ball-type priors, which can handle zeros in data.
- Returners and explorers dichotomy in the face of natural hazards
2024 · Scientific Reports
Examines how people move differently during natural disasters, separating those who return to familiar places from those who explore new areas.
- Spectral Clustering, Bayesian Spanning Forest, and Forest Process
2023 · Journal of the American Statistical Association
Links spectral clustering to a Bayesian forest model to estimate clustering uncertainty.
Plain-language summaries are AI-written and reviewed by us; check the paper for details.
How to reach out
Leo Li Duan hasn't set up this page yet. You can also look them up in the UF directory (search “Leo Li Duan”).