Minseok Choi
Pohang University of Science and Technology · 意思決定科学
研究室紹介
Professor Minseok Choi's research lab specializes in uncertainty quantification and machine learning for scientific computing, with a strong focus on stochastic partial differential equations (SPDEs) and high-dimensional function approximation. The lab develops advanced Bayesian and physics-informed deep learning frameworks—such as Bayesian PINNs and MGDGANs—to model complex systems with inherent randomness and physical constraints. Research also spans computational statistics, polynomial chaos expansions, and social network analysis of scientific collaboration in convergence technologies, reflecting a multidisciplinary approach to innovation in science and engineering.
Research Overview
Research Output Trend
Figures are computed from collected data and may differ slightly.
Selected Papers
15We focus on the analysis of variance (ANOVA) method for high dimensional function approximation using Jacobi polynomial chaos to represent the terms of the expansion. First, we develop a weight theory inspired by quasi-Monte Carlo theory to identify which functions have low effective dimension using the ANOVA expansion in different norms. We then present estimates for the truncation error in the ANOVA expansion and for the interpolation error using multielement polynomial chaos in the weighted K
.Bayesian physics-informed neural networks (B-PINNs) have emerged as an efficient tool for uncertainty quantification in partial differential equations (PDEs). However, their applicability has been limited to accounting for noisy data. They fail to effectively address the uncertainty arising from the randomness of physical parameters in stochastic PDEs (SPDEs). To this end, we propose a novel Bayesian deep learning framework designed for uncertainty quantification in SPDE problems. We model the
We propose novel structures of generator and discriminator in physics-informed generative adversarial networks called multiple-generator-and-discriminator generative adversarial networks (MGDGANs), that are designed to solve stochastic partial differential equations (SPDEs). MGDGANs for SPDEs consist of three steps: a generator that samples a solution to the SPDEs, a physics-informed operator that enforces the governing equation, and a discriminator that distinguishes between samples from the ge
본 연구는 국내 융합기술 분야 연구자들 간의 연구 협력 행태를 알아보기 위하여 사회적 연결망 분석(social network analysis)을 수행하였다. 융합기술분야 연구자 1,095명로 구성된 국내 융합기술 연구자 네트워크의 특성은 다음과 같다. 첫째, 융합기술 연구자들은 다른 과학기술분야에 비해 인당 논문 편수가 많아, 융합기술분야에서 새로운 지식 창조가 활발하게 이루어지고 있음을 보여주고 있다. 둘째, 높은 생산성에 비해 융합기술연구자 그룹의 상당수가 논문의 공동저작을 활발하게 진행하고 있지 않고 제한된 협력 관계만을 보여주고 있다. 마지막으로, 네트워크 구성원들 간의 거리가 다른 과학기술분야의 그것에 비해 가깝지만, 삼각형 모양 관계의 공동 연구하는 형태로 발전하지 않고 양자 간의 공동 연구로만 머물러 있는 경우가 많다. 오직 소수의 연구자들의 허브 역할로 연구자들 간의 연결고리가 유지되고 있다. 즉, 국내 융합기술 연구 분야의 협력 관계는 분권형 이라기보다는 집중형의 특
A three-dimensional computation was conducted to understand effects of the inlet boundary layer thickness on the internal flow in a low-speed axial compressor operating at the design condition() and near stall condition(). At the design condition, the flows in the axial compressor show, independent of the inlet boundary layer thickness, similar characteristics such as the pressure distribution, size of the hub corner-stall, tip leakage flow trajectory, limiting streamlines on the blade suction s
Today convergence technologies have become a major issue in science policy. This paper describes the current state of scientific collaboration in convergence technologies among researchers in South Korea, by conducting survey and the Social Network Analysis (SNA) with a data set of 1,095 researchers who have involved in the development of the convergence technologies. The main research findings are fivefold. First, dominant numbers of researchers are involved in convergence technology with IT be