Robust experimental design
Constructing space-filling designs that remain informative when portions of the input space produce failed or unusable runs.
StatisticsComputer experimentsUncertainty quantification
Sifan Tao
Ph.D. student in Statistics at the University of Wisconsin-Madison
I develop statistical designs and methods for reliable computer experiments, with a focus on Latin hypercube designs, failure-aware experimentation, and multi-model data fusion.
Advised by Professor Peter Chien.
- Based in
- Madison, Wisconsin
01
Research
Designing better experiments for complex computational systems
My work sits at the intersection of experimental design, statistical computation, and uncertainty quantification.
02
Selected work
Publications and manuscripts
Group Representative Latin Hypercube Design
Sifan Tao and Peter Chien
Spatiotemporal Origin and Dynamics of Sodium Plating in Sodium-Ion Batteries
Ziqi Yang, Sifan Tao, Qianli Xing, and Fang Liu
Robust Latin Hypercube Design for Computer Experiments with Potential Failure Regions
Sifan Tao and Peter Chien
Overlapping Sliced Latin Hypercube Designs for Fusing Multiple Computer Models
Xiaoyu Lei*, Sifan Tao*, Peter Chien, and Joshua Cape
* Equal contribution
03
Background
Education and recent experience
Education
2023 to present
Ph.D. in Statistics
University of Wisconsin-Madison
2019 to 2023
B.S. in Statistics
University of Virginia
Fall 2020
UVA at Fudan: Go Local
School of Data Science, Fudan University
Recent experience
Summer 2026
Data Science Ph.D. Intern
Capital One, Upmarket Acquisition Behavioral
Contact
Let’s talk about research.
I welcome conversations about experimental design, computer experiments, and uncertainty quantification.
sifan.tao@wisc.edu