Kyungsik Lee
Seoul National University · 工学
研究室紹介
Professor Kyungsik Lee's research lab specializes in optimization, machine learning, and data science with a focus on developing advanced algorithms for real-world applications in energy systems, finance, and high-dimensional data analysis. The lab emphasizes scalable and interpretable methods, such as sparse and least-angle variants of principal component analysis, variable selection techniques with adaptive regularization, and optimization models for electric vehicle charging and credit scoring. A recurring theme is the integration of mathematical programming with statistical learning to address complex decision-making problems under uncertainty and structural constraints. The lab also explores polyhedral structures in combinatorial optimization, particularly in network and graph-based problems.
Research Overview
Research Output Trend
Figures are computed from collected data and may differ slightly.
Selected Papers
15The large-scale integration of electric vehicles (EVs) into power systems is expected to lead to challenges in the operation of the charging infrastructure. In this paper, we deal with the problem of an aggregator coordinating charging schedules of EVs with the objective of minimizing the total charging cost. In particular, unlike most previous studies, which assumed constant maximum charging power, we assume that the maximum charging power can vary according to the current state of charge (SOC)
We develop several variable selection methods using signomial function to select relevant variables for multi-class classification by taking all classes into consideration. We introduce a -norm regularization function to measure the number of selected variables and two adaptive parameters to apply different importance weights for different variables according to their relative importance. The proposed methods select variables suitable for predicting the output and automatically determine the num
Abstract Principal component analysis (PCA) has been a widely used technique for dimension reduction while retaining essential information. However, the ordinary PCA lacks interpretability, especially when dealing with large scale data. To address this limitation, sparse PCA (SPCA) has emerged as an interpretable variant of ordinary PCA. However, the ordinary SPCA relies on solving a challenging non-convex discrete optimization problem, which maximizes explained variance while constraining the n
The minimum multicut problem on a cycle is to find a multicut on an undirected cycle such that the sum of weights is minimized, which is known to be polynomially solvable. This paper shows that there exists a compact polyhedral description of the set of feasible solutions to the problem whose number of variables and constraints is О(υκ).