Hi! My name is Junhyoung Chung.
I am currently a research intern at Krafton AI. I obtained my M.S. degree in Statistics from Seoul National University, advised by Professor Gunwoong Park. Previously, I obtained my B.S. degree in Statistics and B.A. degree in Economics from the same university.
My academic goal is to develop principled statistical methods that distill essential structures from imperfect and complex data. I aim to bridge the gap between theoretical rigor and practical applicability with the following objectives:
- Developing robust statistical frameworks resilient to data contamination.
- Uncovering latent structures and geometric alignment via optimal transport.
- Establishing theoretical guarantees for reliable machine learning systems.
My research initially focused on probabilistic graphical models, specifically designing consistent structure learning algorithms in the presence of measurement error. Recently, I have expanded my interest to optimal transport to robustly align non-Euclidean data structures. I have also carried out applied research introducing statistical models tailored to real-world problem solving.
News
- Apr 2026 New paper! My work "Convex distance operator transport: A convex and geometry-preserving formulation" has been accepted at ICML 2026. The paper and the code will be released shortly.
- Mar 2026 I started as a research intern at Krafton AI.
- Feb 2026 I received my M.S. degree in Statistics from Seoul National University.
- Dec 2025 I will be attending NeurIPS 2025.
- Aug 2025 New paper! "Discovering causal structures in corrupted data: Frugality in anchored Gaussian DAG models" has been published in Computational Statistics and Data Analysis.
- Aug 2025 New paper! "Prediction of high-risk mountain accident areas using a Hurdle model" has been published in Korean Journal of Applied Statistics (written in Korean).