About me | Minji Kim

Hi, I'm Minji Kim!

I am a Ph.D. candidate in Statistics and Operations Research at UNC Chapel Hill under Prof. Vladas Pipiras.
My research lies in the field of modeling physical systems with computer simulation codes, developing statistical methods for data-efficient learning in multi-fidelity and semi-supervised settings.

How to utilize data efficiently when multiple simulation codes for the same quantity with different fidelities are available?
We develop multi-fidelity strategies for estimating the distribution of high-fidelity outputs, with a particular focus on extreme events. We explore two strategies for leveraging low-fidelity data:
(i) Selective sampling: Low-fidelity model outputs are first evaluated to guide the sampling of high-fidelity outputs. We apply this in a non-parametric setting, incorporating extreme value theory to improve tail estimation.
(ii) Data fusion: A larger amount of offline, independently obtained low-fidelity data is used to construct estimators with reduced variance. We first develop this approach in a parametric setting, then extend it to a non-parametric setting.

I applied my skills during internships at Google (LLM evaluation, 2025), Moloco (real-time bidding for app advertisement, 2023), and LLNL (data-driven acceleration of physics simulations, 2022).

More details are in the Research tab or my Resume.

Ph.D: Statistics and Operations Research, UNC Chapel Hill, May 2026 (expected)
M.S: Statistics, Seoul National University, Feb 2021
B.S: Statistics, Minor in Computer Science and Engineering, Seoul National University, Feb 2019