Juho Lee
Korea Advanced Institute of Science and Technology · 情報科学
研究室紹介
Professor Juho Lee's research lab specializes in machine learning, statistical modeling, and computational neuroscience, with a focus on few-shot and few-data learning, neural processes, and brain-computer interfaces. The lab develops data-driven methods such as transductive learning, bootstrap-based neural processes, and domain adaptation to improve model generalization and reliability in low-data regimes. It also explores neurophysiological applications, including gamma entrainment stimulation and EEG-based motor imagery classification, aiming to bridge machine learning with neuroscience and clinical applications. The lab emphasizes scalable inference, interpretability, and real-world applicability in both artificial intelligence and biomedical systems.
Research Overview
Research Output Trend
Figures are computed from collected data and may differ slightly.
Selected Papers
15Few-shot learning aims to build a learner that quickly generalizes to novel classes even when a limited number of labeled examples (so-called low-data problem) are available. Meta-learning is commonly deployed to mimic the test environment in a training phase for good generalization, where episodes (i.e., learning problems) are manually constructed from the training set. This framework gains a lot of attention to few-shot learning with impressive performance, though the low-data problem is not f
Abstract Introduction The accumulation of amyloid-beta (Aβ) is one of the neuropathologic hallmarks of Alzheimer’s disease (AD) and abnormal gamma band oscillations and brain connectivity have been observed. Recently, a therapeutic potential of gamma entrainment of the brain was reported by Iaccarino et al . However, the affected areas were limited to hippocampus and visual cortex. Therefore, we sought to test the effects of acoustic stimulation in a mouse model of AD. Methods Freely moving 6-mo
Unlike in the traditional statistical modeling for which a user typically hand-specify a prior, Neural Processes (NPs) implicitly define a broad class of stochastic processes with neural networks. Given a data stream, NP learns a stochastic process that best describes the data. While this "data-driven" way of learning stochastic processes has proven to handle various types of data, NPs still rely on an assumption that uncertainty in stochastic processes is modeled by a single latent variable, wh
Discriminating motor imagery with electroencephalogram (EEG)-based brain-computer interface (BCI) poses a challenge as it involves an extensive data acquisition phase that demands a substantial amount of effort from the user. To address this issue, one approach is to use unsupervised domain adaptation, where classification models are constructed using data from multiple subjects, and only the unlabeled data from the target user is used for model calibration. However, since brain patterns from mo
We consider a non-projective class of inhomogeneous random graph models with interpretable parameters and a number of interesting asymptotic properties. Using the results of Bollobás et al. [2007], we show that i) the class of models is sparse and ii) depending on the choice of the parameters, the model is either scale-free, with power-law exponent greater than 2, or with an asymptotic degree distribution which is power-law with exponential cut-off. We propose an extension of the model that can
We examine whether and to what extent unions inhibit labor flexibility in the Korean manufacturing. We provide evidence that the short-run employment and hours adjustment of manufacturing regular workers decrease in the post-1987 period with the abrupt incidence of unleashing active unionism in 1987. However, negative union effects on employment adjustment are limited to male, production, and regular workers. We also note that significant part of the decrease in employment flexibility is attribu
Discriminating motor imagery with electroencephalogram (EEG)-based brain-computer interface (BCI) poses a challenge as it involves an extensive data acquisition phase that demands a substantial amount of effort from the user. To address this issue, one approach is to use unsupervised domain adaptation, where classification models are constructed using data from multiple subjects, and only the unlabeled data from the target user is used for model calibration. However, since brain patterns from mo
최근 들어 푸코의 신자유주의적 통치성 개념이 국내 교육학 내에서 활발하게 연구되고 있음에도 불구하고, 교육철학 분야 내에서는 현재까지 제대로 논의되지 않고 있다. 특히 신자유주의적 통치성 개념은 교육학의 기본 개념인 주체화 개념과 매우 밀접한 연관성을 갖고 있기 때문에, 인간 형성의 문제를 주요 탐구 주제로 삼고 있는 교육철학 분야 내에서 반드시 심도 있게 논의되어야 한다. 이러한 연구의 필요성을 토대로 본 연구는 먼저 푸코의 통치, 자기의 인도·타인의 인도·인도의 인도 및 자기의 기술·지배의 기술 개념을 각각 살펴보면서, 푸코가 제시한 통치성 개념의 핵심적 의미, 즉 통치행위는 자기의 인도를 위한 자기의 기술과 타인의 인도를 위한 지배의 기술이 서로 결합하면서 이루어지고 있다는 점을 명확하게 드러냈다. 또한, 푸코가 분석한 신자유주의적 통치성의 사례로서의 인적 자본론과 기업가적 자아 개념을 토대로 본 연구는 신자유주의의 사회문화적 조건 내에서 존재하는 지배의 기술과 자기의 기술이
Lee, Ju-Ho.November, 2015.Innovation in national R&D program,Proceedings,KDI Schoolof Public Policy and Management,49
<p>Discriminating motor imagery with electroencephalogram (EEG)-based brain-computer interface (BCI) poses a challenge as it involves an extensive data acquisition phase that demands a substantial amount of effort from the user. To address this issue, one approach is to use unsupervised domain adaptation, where classification models are constructed using data from multiple subjects, and only the unlabeled data from the target user is used for model calibration. However, since brain pattern
This paper deals with the capital budgeting problem of a firm where investments are risky and interrelated. The established models might be classified into two categories; One is the chance-constrained programming model and the other is the expected utility maximization model. The former has a rather limited objective function and does not consider the risk in direct manner. The latter, on the other hand, might lead to a wrong decision because it uses an approximate value of expected utility. Th