Jae-Hun Jung
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Jae-Hun Jung has a total of 98 co-authors. Here are the top co-authors by number of shared publications:
- Jiaxi Gu
Department of Mathematics & POSTECH MINDS (Mathematical INstitute for Data Science), Pohang University of Science and Technology, Pohang, Korea
10 shared publications - Xinjuan Chen
Department of Mathematics, College of Science, Jimei University, Xiamen, China
7 shared publications - Dongjin Lee
Pohang University of Science and Technology (POSTECH), Pohang 37673, Republic of Korea
5 shared publications Eunwoo Heo
Pohang University of Science and Technology
5 shared publications- Kwanghyuk Park
Pohang University of Science and Technology
5 shared publications - Daniel Olmos-Liceaga
Universidad de Sonora, México
4 shared publications
Latest Publications
- 29 September 2026
- Published by Taylor & Francis in Journal of New Music Research
- p. 1-25
A diffusion model for generating unknown Pinax data of Athanasius Kircher's Arca Musarithmica
1. Automatic music generation has a long history. One significant example from the 1600s is Athanasius Kircher's concept of a music-generating machine (Gouk, 2001; Nierhaus, 2009). In his work Musu...
- 29 September 2026
- Published by Taylor & Francis in Journal of New Music Research
- p. 1-25
A diffusion model for generating unknown Pinax data of Athanasius Kircher's Arca Musarithmica
1. Automatic music generation has a long history. One significant example from the 1600s is Athanasius Kircher's concept of a music-generating machine (Gouk, 2001; Nierhaus, 2009). In his work Musu...
- 1 September 2026
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 36 (9)
Zipf–Mandelbrot scaling in Korean court music: Universal patterns in music
Rank–frequency distributions in many complex systems, including natural language, are often well described by the Zipf–Mandelbrot law. Due to the semantic characteristics of this law, it has also been observed in musical data. However, most such studies have focused on Western music, and its applicability to non-Western music remains not well investigated. We analyzed 43 Korean court music pieces called Jeong-ak, spanning several centuries and written in the traditional Korean musical notation Jeongganbo. These pieces were transcribed into Western staff notation, and musical data such as pitch and duration were extracted. Using pitch, duration, and their paired combinations as Zipfian units, we found that Korean music also fits the Zipf–Mandelbrot law to a high degree, particularly for the paired pitch–duration unit. Korean music has evolved collectively over long periods, smoothing idiosyncratic variations and producing forms that are widely understandable among people. This collective evolution appears to have played a significant role in shaping the characteristics that lead to the satisfaction of Zipf–Mandelbrot law. Our findings provide additional evidence that Zipf–Mandelbrot scaling in musical data is universal across cultures. We further show that the joint distribution of two independent Zipfian data sets follows the Zipf–Mandelbrot law.
- 1 September 2026
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 36 (9)
Zipf–Mandelbrot scaling in Korean court music: Universal patterns in music
Rank–frequency distributions in many complex systems, including natural language, are often well described by the Zipf–Mandelbrot law. Due to the semantic characteristics of this law, it has also been observed in musical data. However, most such studies have focused on Western music, and its applicability to non-Western music remains not well investigated. We analyzed 43 Korean court music pieces called Jeong-ak, spanning several centuries and written in the traditional Korean musical notation Jeongganbo. These pieces were transcribed into Western staff notation, and musical data such as pitch and duration were extracted. Using pitch, duration, and their paired combinations as Zipfian units, we found that Korean music also fits the Zipf–Mandelbrot law to a high degree, particularly for the paired pitch–duration unit. Korean music has evolved collectively over long periods, smoothing idiosyncratic variations and producing forms that are widely understandable among people. This collective evolution appears to have played a significant role in shaping the characteristics that lead to the satisfaction of Zipf–Mandelbrot law. Our findings provide additional evidence that Zipf–Mandelbrot scaling in musical data is universal across cultures. We further show that the joint distribution of two independent Zipfian data sets follows the Zipf–Mandelbrot law.
- Open Access
- 11 June 2026
- Published by Springer Nature in Advances in Data Analysis and Classification
- p. 1-49
Weighted isolation and random cut forests for anomaly detection
Random cut forest (RCF) algorithms have been developed for anomaly detection, particularly in time series data. The RCF algorithm is an improved version of the isolation forest (IF) algorithm. Unlike the IF algorithm, the RCF algorithm can determine whether real-time input contains an anomaly by inserting the input into the constructed tree network. Various RCF algorithms, including Robust RCF (RRCF), have been developed, where the cutting procedure is adaptively chosen probabilistically. The RRCF algorithm demonstrates better performance than the IF algorithm, as dimension cuts are decided based on the geometric range of the data, whereas the IF algorithm randomly chooses dimension cuts. However, the overall data structure is not considered in both IF and RRCF, given that split values are chosen randomly. In this paper, we propose new IF and RCF algorithms, referred to as the weighted IF (WIF) and weighted RCF (WRCF) algorithms, respectively. Their split values are determined by considering the density of the given data. To introduce the WIF and WRCF, we first present a new geometric measure, a density measure , which is crucial for constructing the WIF and WRCF. We provide various mathematical properties of the density measure, accompanied by theorems that support and validate our claims through numerical examples. As the tree algorithms rely on geometric analysis, we first provide numerical examples for geometric data sets. Then we apply the WRCF to time series data sets and demonstrate that the proposed method performs better than the existing RRCF.
- 11 June 2026
- Published by Springer Nature in Advances in Data Analysis and Classification
- p. 1-49
Open AccessWeighted isolation and random cut forests for anomaly detection
Random cut forest (RCF) algorithms have been developed for anomaly detection, particularly in time series data. The RCF algorithm is an improved version of the isolation forest (IF) algorithm. Unlike the IF algorithm, the RCF algorithm can determine whether real-time input contains an anomaly by inserting the input into the constructed tree network. Various RCF algorithms, including Robust RCF (RRCF), have been developed, where the cutting procedure is adaptively chosen probabilistically. The RRCF algorithm demonstrates better performance than the IF algorithm, as dimension cuts are decided based on the geometric range of the data, whereas the IF algorithm randomly chooses dimension cuts. However, the overall data structure is not considered in both IF and RRCF, given that split values are chosen randomly. In this paper, we propose new IF and RCF algorithms, referred to as the weighted IF (WIF) and weighted RCF (WRCF) algorithms, respectively. Their split values are determined by considering the density of the given data. To introduce the WIF and WRCF, we first present a new geometric measure, a density measure , which is crucial for constructing the WIF and WRCF. We provide various mathematical properties of the density measure, accompanied by theorems that support and validate our claims through numerical examples. As the tree algorithms rely on geometric analysis, we first provide numerical examples for geometric data sets. Then we apply the WRCF to time series data sets and demonstrate that the proposed method performs better than the existing RRCF.
- 20 May 2026
- Published by Elsevier in Infant Behavior and Development
- Vol. 83, 102198
Cross-domain evaluation and fine-tuned adaptation of iCatcher+ for Korean infant gaze data
- 20 May 2026
- Published by Elsevier in Infant Behavior and Development
- Vol. 83, 102198
Cross-domain evaluation and fine-tuned adaptation of iCatcher+ for Korean infant gaze data
- Open Access
- 30 March 2026
- Published by IOP Publishing in Machine Learning: Science and Technology
- Vol. 7 (2) , 025035
Drop-in density-aware partitioning for tree-based anomaly detection
Unsupervised isolation tree-based anomaly detection methods are widely used for their simplicity, scalability, and strong empirical performance. However, the standard partition rule, whether drawing thresholds uniformly from feature ranges or sampling random normal vectors for oblique hyperplanes, ignores the underlying data distribution and often cuts through dense clusters rather than sparse regions. To address this limitation, we introduce density-aware partitioning (DAP), a unified framework that guides data partitioning according to local data density. DAP consists of two components: density-aware split, which replaces uniform threshold sampling with a density-aware rejection sampler guided by a counting map, and density-aware direction, which samples a normal vector for hyperplanes that tend to separate low-density gaps rather than dense clusters. DAP can be integrated into various isolation tree-based models, enhancing anomaly score discrimination with minimal runtime overhead, as the simplified tree structure compensates for the slight increase in construction time. Through synthetic visualizations and large-scale benchmark evaluations, we demonstrate that DAP consistently improves detection accuracy without incurring significant computational overhead. Moreover, we theoretically analyze the counting map and establish its key properties, providing a formal justification for the tolerance parameter and the efficiency of the rejection mechanism. Overall, DAP offers a simple yet effective enhancement that improves partitioning capability without altering the core design of isolation tree-based models.
- 30 March 2026
- Published by IOP Publishing in Machine Learning: Science and Technology
- Vol. 7 (2) , 025035
Open AccessDrop-in density-aware partitioning for tree-based anomaly detection
Unsupervised isolation tree-based anomaly detection methods are widely used for their simplicity, scalability, and strong empirical performance. However, the standard partition rule, whether drawing thresholds uniformly from feature ranges or sampling random normal vectors for oblique hyperplanes, ignores the underlying data distribution and often cuts through dense clusters rather than sparse regions. To address this limitation, we introduce density-aware partitioning (DAP), a unified framework that guides data partitioning according to local data density. DAP consists of two components: density-aware split, which replaces uniform threshold sampling with a density-aware rejection sampler guided by a counting map, and density-aware direction, which samples a normal vector for hyperplanes that tend to separate low-density gaps rather than dense clusters. DAP can be integrated into various isolation tree-based models, enhancing anomaly score discrimination with minimal runtime overhead, as the simplified tree structure compensates for the slight increase in construction time. Through synthetic visualizations and large-scale benchmark evaluations, we demonstrate that DAP consistently improves detection accuracy without incurring significant computational overhead. Moreover, we theoretically analyze the counting map and establish its key properties, providing a formal justification for the tolerance parameter and the efficiency of the rejection mechanism. Overall, DAP offers a simple yet effective enhancement that improves partitioning capability without altering the core design of isolation tree-based models.
- Open Access
- 9 March 2026
- Published by Elsevier in SSRN Electronic Journal
Cross-Domain Evaluation and Fine-Tuned Adaptation of iCatcher+ for Korean Infant Gaze Data
This study evaluates the cross-domain generalisability of three iCatcher+ gaze classification models on a dataset of Korean infants. Applying the pretrained models to the Korean dataset resulted in a substantial decline in accuracy across all models. Performance was systematically modulated by developmental mismatch: accuracy declined as the age distribution of the target dataset diverged from that of the training data. In contrast, experiment-level factors such as stimulus type and temporal progression had limited effects. Fine-tuning the Lookit model on Korean data improved target-domain performance, but at the cost of reduced accuracy on the original source dataset, consistent with catastrophic forgetting. Our findings highlight that deep learning-based gaze classifiers remain sensitive to domain shift. Although fine-tuning can partially mitigate performance loss, robust generalisation requires closer alignment between training and deployment contexts.
- 9 March 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessCross-Domain Evaluation and Fine-Tuned Adaptation of iCatcher+ for Korean Infant Gaze Data
This study evaluates the cross-domain generalisability of three iCatcher+ gaze classification models on a dataset of Korean infants. Applying the pretrained models to the Korean dataset resulted in a substantial decline in accuracy across all models. Performance was systematically modulated by developmental mismatch: accuracy declined as the age distribution of the target dataset diverged from that of the training data. In contrast, experiment-level factors such as stimulus type and temporal progression had limited effects. Fine-tuning the Lookit model on Korean data improved target-domain performance, but at the cost of reduced accuracy on the original source dataset, consistent with catastrophic forgetting. Our findings highlight that deep learning-based gaze classifiers remain sensitive to domain shift. Although fine-tuning can partially mitigate performance loss, robust generalisation requires closer alignment between training and deployment contexts.
- Open Access
Zipf-Mandelbrot Scaling in Korean Court Music: Universal Patterns in Music
Zipf's law, originally discovered in natural language and later generalized to the Zipf-Mandelbrot law, describes a power-law relationship between the frequency of a Zipfian element and its rank. Due to the semantic characteristics of this law, it has also been observed in musical data. However, most such studies have focused on Western music, and its applicability to non-Western music remains not well investigated. We analyzed 43 Korean court music pieces called Jeong-ak, spanning several centuries and written in the traditional Korean musical notation Jeongganbo. These pieces were transcribed into Western staff notation, and musical data such as pitch and duration were extracted. Using pitch, duration, and their paired combinations as Zipfian units, we found that Korean music also fits the Zipf-Mandelbrot law to a high degree, particularly for the paired pitch-duration unit. Korean music has evolved collectively over long periods, smoothing idiosyncratic variations and producing forms that are widely understandable among people. This collective evolution appears to have played a significant role in shaping the characteristics that lead to the satisfaction of Zipf-Mandelbrot law. Our findings provide additional evidence that Zipf-Mandelbrot scaling in musical data is universal across cultures. We further show that the joint distribution of two independent Zipfian data sets follows the Zipf-Mandelbrot law; in this sense, our result does not merely extend Zipf's law but deepens our understanding of how scaling laws behave under composition and interaction, offering a more unified perspective on rank-based statistical regularities.
Open AccessZipf-Mandelbrot Scaling in Korean Court Music: Universal Patterns in Music
Zipf's law, originally discovered in natural language and later generalized to the Zipf-Mandelbrot law, describes a power-law relationship between the frequency of a Zipfian element and its rank. Due to the semantic characteristics of this law, it has also been observed in musical data. However, most such studies have focused on Western music, and its applicability to non-Western music remains not well investigated. We analyzed 43 Korean court music pieces called Jeong-ak, spanning several centuries and written in the traditional Korean musical notation Jeongganbo. These pieces were transcribed into Western staff notation, and musical data such as pitch and duration were extracted. Using pitch, duration, and their paired combinations as Zipfian units, we found that Korean music also fits the Zipf-Mandelbrot law to a high degree, particularly for the paired pitch-duration unit. Korean music has evolved collectively over long periods, smoothing idiosyncratic variations and producing forms that are widely understandable among people. This collective evolution appears to have played a significant role in shaping the characteristics that lead to the satisfaction of Zipf-Mandelbrot law. Our findings provide additional evidence that Zipf-Mandelbrot scaling in musical data is universal across cultures. We further show that the joint distribution of two independent Zipfian data sets follows the Zipf-Mandelbrot law; in this sense, our result does not merely extend Zipf's law but deepens our understanding of how scaling laws behave under composition and interaction, offering a more unified perspective on rank-based statistical regularities.
- 23 December 2025
- Published by Elsevier in Advances in Applied Mathematics
- Vol. 174
Exact multi-parameter persistent homology of time-series data: Fast and variable topological inferences
- 23 December 2025
- Published by Elsevier in Advances in Applied Mathematics
- Vol. 174
Exact multi-parameter persistent homology of time-series data: Fast and variable topological inferences
- Open Access
- 21 November 2025
- Published by American Astronomical Society in The Astronomical Journal
- Vol. 170 (6) , 337
Investigating the Effects of Point Source Injection Strategies on KMTNet Real/Bogus Classification
Recently, machine learning-based real / bogus (RB) classifiers have demonstrated effectiveness in filtering out artifacts and identifying genuine transients in real-time astronomical surveys. However, the rarity of transient events and the extensive human labeling required for a large number of samples pose significant challenges in constructing training data sets for RB classification. Given these challenges, point source injection techniques, which inject simulated point sources into optical images, provide a promising solution. This paper presents the first detailed comparison of different point source injection strategies and their effects on classification performance within a simulation-to-reality framework. To this end, we first construct various training data sets based on random injection (RI), near galaxy injection (NGI), and a combined approach by using the Korea Microlensing Telescope Network data sets. Subsequently, we train convolutional neural networks on simulated cutout samples and evaluate them on real, imbalanced data sets from gravitational wave follow-up observations for GW190814 and S230518h. Extensive experimental results show that RI excels at asteroid detection and bogus filtering but underperforms on transients occurring near galaxies (e.g., supernovae). In contrast, NGI is effective for detecting transients near galaxies but tends to misclassify variable stars as transients, resulting in a high false positive rate. The combined approach effectively handles these trade-offs, thereby balancing between detection rate and false positive rate. Our results emphasize the importance of point source injection strategy in developing robust RB classifiers for transient (or multimessenger) follow-up campaigns.
- 21 November 2025
- Published by American Astronomical Society in The Astronomical Journal
- Vol. 170 (6) , 337
Open AccessInvestigating the Effects of Point Source Injection Strategies on KMTNet Real/Bogus Classification
Recently, machine learning-based real / bogus (RB) classifiers have demonstrated effectiveness in filtering out artifacts and identifying genuine transients in real-time astronomical surveys. However, the rarity of transient events and the extensive human labeling required for a large number of samples pose significant challenges in constructing training data sets for RB classification. Given these challenges, point source injection techniques, which inject simulated point sources into optical images, provide a promising solution. This paper presents the first detailed comparison of different point source injection strategies and their effects on classification performance within a simulation-to-reality framework. To this end, we first construct various training data sets based on random injection (RI), near galaxy injection (NGI), and a combined approach by using the Korea Microlensing Telescope Network data sets. Subsequently, we train convolutional neural networks on simulated cutout samples and evaluate them on real, imbalanced data sets from gravitational wave follow-up observations for GW190814 and S230518h. Extensive experimental results show that RI excels at asteroid detection and bogus filtering but underperforms on transients occurring near galaxies (e.g., supernovae). In contrast, NGI is effective for detecting transients near galaxies but tends to misclassify variable stars as transients, resulting in a high false positive rate. The combined approach effectively handles these trade-offs, thereby balancing between detection rate and false positive rate. Our results emphasize the importance of point source injection strategy in developing robust RB classifiers for transient (or multimessenger) follow-up campaigns.
- 21 November 2025
- Published by Elsevier in Neurocomputing
- Vol. 665
PHLP: Sole persistent homology for link prediction - interpretable feature extraction
- 21 November 2025
- Published by Elsevier in Neurocomputing
- Vol. 665
PHLP: Sole persistent homology for link prediction - interpretable feature extraction