I am an associate professor in the Department of Statistics at Jeonbuk National University. I received my Ph.D. in 2018 from the Department of Statistics at Seoul National University, under the supervision of Professor Hee-Seok Oh. My research area is multiscale methods, which aims to represent and analyze data from multiple levels or scales. I find multiscale thinking fascinating because it allows us to observe the same data from different perspectives, often leading to new insights. Sometimes, changing the scale of observation can even reverse the conclusion—an effect well illustrated by Simpson’s paradox. The idea of using hierarchical observation tools to interpret data reveals the richness and complexity of statistical reasoning, and that’s what makes multiscale methods so interesting to me.
More recently, I’ve been focusing on multiscale methods for non-Euclidean data, such as manifolds, graphs, simplicial complexes, and other combinatorial representations. While classical multiscale techniques—like Fourier and wavelet transforms—are mostly developed for Euclidean domains, the underlying idea of multiscale analysis is much broader. It is not tied to any specific data structure, but rather to how we observe and decompose information across levels. That’s why developing multiscale methods for non-Euclidean domains is not just a technical extension, but a deeper exploration of what multiscale analysis really means. For me, this is the most exciting part—it pushes us to rethink the foundations of multiscale thinking itself. As a bonus, this line of research also aligns with emerging fields like geometric deep learning and topological deep learning, where non-Euclidean structures are naturally encountered. Comparing their approaches with mine is one of the things I truly enjoy.
- Position: Chair, Department of Statistics | Director, Applied Statistics Research Institute
- Email: guebin@jbnu.ac.kr
- Phone: (+82) 63-270-3387
- YouTube: https://www.youtube.com/channel/UCQk9RyBNgXc7ORIsYlOfQrg
This section introduces only the research projects I personally led. Collaborative projects where I was not the Principal Investigator (PI) are excluded.
- Geometric Deep Learning: Statistical Methodology for Non-Euclidean Data (생애 첫 연구, 2021.09 ~ 2022.08, 2021R1G1A1094937)
- A Study of Multiscale Methods in Non-Euclidean Data (지역대학우수과학자, 2023.06 ~ 2029.05, RS-2023-00249743)
This section highlights select research accomplishments that directly align with my core research.
- Choi, S., & Choi, G. (2025). Gode: graph Fourier transform based outlier detection using empirical Bayesian thresholding. Journal of the Korean Statistical Society, 1-21.
- Choi, G., & Oh, H. S. (2024). Decomposition via elastic-band transform. Pattern Recognition Letters, 182, 76-82.
- Choi, G., & Oh, H. S. (2023). Elastic-band transform for visualization and detection. Pattern Recognition Letters, 166, 119-125.
- Kim, D., Choi, G., & Oh, H. S. (2020). Ensemble patch transformation: a flexible frame- work for decomposition and filtering of signal. EURASIP Journal on Advances in Signal Processing, 2020(1), 1-27.
- Choi, G., Oh, H. S. & Kim, D. (2018). Enhancement of variational mode decomposition with missing values. Signal Processing, 142, 75–86.
For a complete list of publications, please click here.
Note: Course names in bold italics are graduate classes
Show older lectures
- Statistics Seminar
- Machine Learning in Practice
- Python Programming
- Statistical Computing
- Deep Learning
- Winter Program: Statistical Data Analysis using Python
- Winter Program: Data Science Toolbox
- Special Topics in Big Data Analysis
- Machine Learning in Practice
- Data Visualization
- Short Lecture: Introduction to Reinforcement Learning
- Summer Program: Statistical Data Analysis using Python
- Advanced Probability Theory
- Stochastic Processes
- Python Programming
- Winter Program: Statistical Data Analysis using Python
- Special Topics in Machine Learning
- Data Visualization
- Deep Learning
- Summer Program: Reinforcement Learning
- Special Topics in Big Data Analysis
- Data Science
- Statistical Computing
- Introduction to Python
- Winter Program: ADsP, Advanced Data analytics semi-Professional
- Winter Program: Introduction to Python
- Special Topics in Data Visualization
- Big Data Analysis
- Data Visualization
- Introduction to R
- Special Topics in Statistical Computing
- Data Science
- Introduction to Python
- Mathematical Statistics
- Time Series Analysis
| Package Name | Documents | Source |
|---|---|---|
| EPT (R) | Paper, CRAN | CRAN |
| EBT (R) | N/A | GitHub |
| graft (Python) | N/A | GitHub |
| gglite (R) | N/A | GitHub |
| gglitely (Python) | N/A | GitHub (under development) |
- 전재범 (2025.03 -)
- N/A
- 이상민 (2025.01 - )