YoonHaeng Hur

Bio

I am a Postdoctoral Research Scientist in the Department of Statistics at Columbia University, supervised by Genevera Allen. I received my PhD in Statistics at the University of Chicago, advised by Tengyuan Liang and Yuehaw Khoo. Before graduate school, I received a BS in Mathematical Sciences and a BA in Economics from Seoul National University.

Research Interests

My research focuses on developing mathematical and statistical foundations for modern data science, using optimal transport as a primary guiding principle. Currently, I am interested in the following topics.

Generative AI via optimal transport and scientific computing
Statistical inference frameworks for reliable unsupervised learning
Scalable data-driven approaches to causal machine learning

Contact: yoonhaeng.hur@columbia.edu

Working Papers

A Convexified Matching Approach to Imputation and Individualized Inference
with T. Liang
[arXiv] [Code]
Learning When the Concept Shifts: Confounding, Invariance, and Dimension Reduction
with K. Dharmakeerthi, T. Liang
[arXiv]

Publications

Robust Point Matching with Distance Profiles
with Y. Khoo
To appear in Journal of Machine Learning Research (2025)
Detecting Weak Distribution Shifts via Displacement Interpolation
with T. Liang
Journal of Business & Economic Statistics (2025)
Reversible Gromov-Monge Sampler for Simulation-Based Inference
with W. Guo, T. Liang
SIAM Journal on Mathematics of Data Science (2024)
Generative Modeling via Tree Tensor Network States
with X. Tang, Y. Khoo, L. Ying
Research in the Mathematical Sciences (2023)
Generative Modeling via Tensor Train Sketching
with J. G. Hoskins, M. Lindsey, E.M. Stoudenmire, Y. Khoo
Applied and Computational Harmonic Analysis (2023) [Code]
Online Learning to Transport via the Minimal Selection Principle
with W. Guo, T. Liang, C.T. Ryan
Conference on Learning Theory (2022)

Bio

Research Interests

Working Papers

Publications

Notes