T-ASE
Knowledge-Assisted Multi-Graph Dependency Learning for Multivariate Time Series Anomaly Detection in Multi-Stage Industrial Processes
IEEE Transactions on Automation Science and Engineering, 2026
Hi, I’m Jaeyeong Lee.
I am a Ph.D. candidate in Industrial & Systems Engineering at KAIST, advised by Prof. Heeyoung Kim. My research lies at the intersection of statistics and machine learning, with a broad goal of developing statistical foundations of learning in function space. I am particularly interested in Bayesian deep learning and Bayesian nonparametrics from a function-space perspective. I also study stochastic systems, including temporal and spatio-temporal dynamics, with applications to industrial systems.
Selected Publications
For a complete list of publications, see Google Scholar ↗.
Journal Articles
Conference Proceedings
UAI
Nonlocal Bayesian Modeling of Continuous Spatio-Temporal Dynamics
Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence (UAI 2026), PMLR 337:3355–3375.
CIKM
GT-PSSM: Unified Probabilistic Framework for Stochastic Dynamics Modeling and Dependency Learning in Multivariate Time Series Anomaly Detection
35th International ACM Conference on Knowledge and Information Management (CIKM 2026), accepted.
Work in Preparation
PREP.
Function-Space Priors for Bayesian Neural Networks via Entropic Lifting and PAC-Bayes Selection
In preparation for submission to ICLR 2027.
Education
Feb. 2024 – Present
KAIST
Ph.D. Candidate in Industrial & Systems Engineering
Advisor: Prof. Heeyoung Kim
Mar. 2022 – Feb. 2024
KAIST
M.S. in Industrial & Systems Engineering
Advisor: Prof. Heeyoung Kim
Mar. 2016 – Feb. 2022
KAIST
B.S. in Mathematical Sciences / Computer Science
Research
Bayesian Deep Learning
Bayesian Neural Networks · Function-Space Priors · Uncertainty Quantification · Bayesian Learning Theory
Bayesian Nonparametrics
Gaussian Processes · Nonparametric Priors · Posterior Contraction · Asymptotic Theory
Stochastic Systems
Stochastic Processes · Time Series · Spatio-Temporal Processes · Continuous-Time Dynamics
Talks & Presentations
International
- J. Lee, T. Yoon, W. Koo, and H. Kim, “Knowledge-Assisted Multi-Graph Structure Learning for Multivariate Time Series Anomaly Detection in Multi-Stage Industrial Processes,” INFORMS Annual Meeting, Atlanta, Georgia, Oct. 26–29, 2025. (Invited Talk)
Domestic
- 이재영, 김희영, “불규칙 시공간 데이터 예측을 위한 베이지안 연속 동적 모델,” 대한산업공학회 추계학술대회, 대전, 2025.
- 이재영, 구원모, 김희영, “다단계 산업 공정 이상탐지를 위한 지식 보조 다중 그래프 구조 학습,” 대한산업공학회 추계학술대회, 서울, 2024.