Daisuke Sakurai
Research Subjects
Research Interests
Daisuke Sakurai has a total of 33 co-authors. Here are the top co-authors by number of shared publications:
- Naoki Hamada
KLab Inc., Tokyo, Japan
3 shared publications Daichi Yamazaki
KLab Inc
2 shared publications- Likun Liu
School of Cyberspace Science, Harbin Institute of Technology, Harbin, China
2 shared publications - Likun Liu
Kyushu University, Fukuoka, Japan
2 shared publications - Ryosuke Ota
Kyushu University, Fukuoka, Japan
2 shared publications - Taiga Yamamoto
Kyushu University
2 shared publications
Latest Publications
- Open Access
- 28 August 2025
- Published by Wiley in Computer Graphics Forum
- Vol. 44 (5)
Arrange and Traverse Algorithm for Computation of Reeb Spaces of Piecewise Linear Maps
We present the first combinatorial algorithm for efficiently computing the Reeb space in all dimensions. The Reeb space is a higher‐dimensional generalization of the Reeb graph, which is standard practice in the analysis of scalar fields, along with other computational topology tools such as persistent homology and the Morse‐Smale complex. One significant limitation of topological tools for scalar fields is that data often involves multiple variables, where joint analysis is more insightful. Generalizing topological data structures to multivariate data has proven challenging and the Reeb space is one of the few available options. However, none of the existing algorithms can efficiently compute the Reeb space in arbitrary dimensions and there are no available implementations which are robust with respect to numerical errors. We propose a new algorithm for computing the Reeb space of a generic piecewise linear map over a simplicial mesh of any dimension called arrange and traverse. We implement a robust specialization of our algorithm for tetrahedral meshes and evaluate it on real‐life data.
- 28 August 2025
- Published by Wiley in Computer Graphics Forum
- Vol. 44 (5)
Open AccessArrange and Traverse Algorithm for Computation of Reeb Spaces of Piecewise Linear Maps
We present the first combinatorial algorithm for efficiently computing the Reeb space in all dimensions. The Reeb space is a higher‐dimensional generalization of the Reeb graph, which is standard practice in the analysis of scalar fields, along with other computational topology tools such as persistent homology and the Morse‐Smale complex. One significant limitation of topological tools for scalar fields is that data often involves multiple variables, where joint analysis is more insightful. Generalizing topological data structures to multivariate data has proven challenging and the Reeb space is one of the few available options. However, none of the existing algorithms can efficiently compute the Reeb space in arbitrary dimensions and there are no available implementations which are robust with respect to numerical errors. We propose a new algorithm for computing the Reeb space of a generic piecewise linear map over a simplicial mesh of any dimension called arrange and traverse. We implement a robust specialization of our algorithm for tetrahedral meshes and evaluate it on real‐life data.
- Open Access
- 14 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 2294-2300
Towards Benchmarking Multi-Objective Optimization Algorithms Based on the Basin Connectivity
- 14 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 2294-2300
Open AccessTowards Benchmarking Multi-Objective Optimization Algorithms Based on the Basin Connectivity
- Open Access
- 14 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 347-350
Visualization of Multiobjective Multimodal Benchmarking Based on Basin Connectivity
Visualization is useful for understanding the behavior of Multiobjective Evolutionary Algorithms (MOEAs). However, it has been challenging to investigate how an MOEA may be trapped by local Pareto optima, in which case some individuals stay inside a "valley" (also known as basins of attraction). To address this, we propose a novel visualization that projects the high-dimensional basins in a 2D plot, in which the individuals of benchmarked MOEAs for user-specified generations are visible. For this, we target specialized benchmark problems known as the Benchmarking Based on Basin Connectivity (3BC), which explicitly models basins of attraction and thus allows us to precompute the necessary geometry as what is called the Reeb space. Our demonstration shows that the proposed visualization displays basins in benchmark problems as intended and, reveals the differences in MOEAs' behaviors in response to the basins. The source code of our system is publically available at https://github.com/dsakurai/benchmark-visualizer.
- 14 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 347-350
Open AccessVisualization of Multiobjective Multimodal Benchmarking Based on Basin Connectivity
Visualization is useful for understanding the behavior of Multiobjective Evolutionary Algorithms (MOEAs). However, it has been challenging to investigate how an MOEA may be trapped by local Pareto optima, in which case some individuals stay inside a "valley" (also known as basins of attraction). To address this, we propose a novel visualization that projects the high-dimensional basins in a 2D plot, in which the individuals of benchmarked MOEAs for user-specified generations are visible. For this, we target specialized benchmark problems known as the Benchmarking Based on Basin Connectivity (3BC), which explicitly models basins of attraction and thus allows us to precompute the necessary geometry as what is called the Reeb space. Our demonstration shows that the proposed visualization displays basins in benchmark problems as intended and, reveals the differences in MOEAs' behaviors in response to the basins. The source code of our system is publically available at https://github.com/dsakurai/benchmark-visualizer.
- Open Access
- 26 June 2023
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 37 (4) , 5266-5275
GenéLive! Generating Rhythm Actions in Love Live!
This article presents our generative model for rhythm action games together with applications in business operation. Rhythm action games are video games in which the player is challenged to issue commands at the right timings during a music session. The timings are rendered in the chart, which consists of visual symbols, called notes, flying through the screen. We introduce our deep generative model, GenéLive!, which outperforms the state-of-the-art model by taking into account musical structures through beats and temporal scales. Thanks to its favorable performance, GenéLive! was put into operation at KLab Inc., a Japan-based video game developer, and reduced the business cost of chart generation by as much as half. The application target included the phenomenal "Love Live!", which has more than 10 million users across Asia and beyond, and is one of the few rhythm action franchises that has led the online era of the genre. In this article, we evaluate the generative performance of GenéLive! using production datasets at KLab as well as open datasets for reproducibility, while the model continues to operate in their business. Our code and the model, tuned and trained using a supercomputer, are publicly available.
- 26 June 2023
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 37 (4) , 5266-5275
Open AccessGenéLive! Generating Rhythm Actions in Love Live!
This article presents our generative model for rhythm action games together with applications in business operation. Rhythm action games are video games in which the player is challenged to issue commands at the right timings during a music session. The timings are rendered in the chart, which consists of visual symbols, called notes, flying through the screen. We introduce our deep generative model, GenéLive!, which outperforms the state-of-the-art model by taking into account musical structures through beats and temporal scales. Thanks to its favorable performance, GenéLive! was put into operation at KLab Inc., a Japan-based video game developer, and reduced the business cost of chart generation by as much as half. The application target included the phenomenal "Love Live!", which has more than 10 million users across Asia and beyond, and is one of the few rhythm action franchises that has led the online era of the genre. In this article, we evaluate the generative performance of GenéLive! using production datasets at KLab as well as open datasets for reproducibility, while the model continues to operate in their business. Our code and the model, tuned and trained using a supercomputer, are publicly available.
- 18 April 2023
- Published by IOP Publishing in Plasma Physics and Controlled Fusion
- Vol. 65 (6) , 065001
Phase jump detection and correction based on the support vector machine
In general, interferometers are used to perform electron density measurements in magnetically confined plasma, where the electron density is dependent on the refractive index of the plasma. Measurements can be made through comparisons of the phase shift variation between the probe and reference laser beam. The plasma electron density should vary continuously during discharge; however, the fringe jump is a step-like change of the apparent electron density caused by a sudden jump of the measured phase shift. The appearance of fringe jump will degrade the interferometric measurements accuracy. This study attempted to solve the fringe jump problem on the polarimeter-interferometer (POINT) diagnostics system of the Experiment Advanced Superconducting Tokamak (EAST) by proposing a support vector machine model for electron density fringe jump detection and correction. The established model can efficiently classify the fringe jump data from the raw measurement data in a manner robust to noise and interference, and subsequently correct the jump. This model greatly improves the correction efficiency and precision of electron density data from the POINT system, and is expected to be embedded into the plasma control system to perform more accurate real-time electron density feedback control. Moreover, the algorithm is not limited to specific fusion devices or interferometer diagnostics, and is applicable to other interferometric measurement systems.
- 18 April 2023
- Published by IOP Publishing in Plasma Physics and Controlled Fusion
- Vol. 65 (6) , 065001
Phase jump detection and correction based on the support vector machine
In general, interferometers are used to perform electron density measurements in magnetically confined plasma, where the electron density is dependent on the refractive index of the plasma. Measurements can be made through comparisons of the phase shift variation between the probe and reference laser beam. The plasma electron density should vary continuously during discharge; however, the fringe jump is a step-like change of the apparent electron density caused by a sudden jump of the measured phase shift. The appearance of fringe jump will degrade the interferometric measurements accuracy. This study attempted to solve the fringe jump problem on the polarimeter-interferometer (POINT) diagnostics system of the Experiment Advanced Superconducting Tokamak (EAST) by proposing a support vector machine model for electron density fringe jump detection and correction. The established model can efficiently classify the fringe jump data from the raw measurement data in a manner robust to noise and interference, and subsequently correct the jump. This model greatly improves the correction efficiency and precision of electron density data from the POINT system, and is expected to be embedded into the plasma control system to perform more accurate real-time electron density feedback control. Moreover, the algorithm is not limited to specific fusion devices or interferometer diagnostics, and is applicable to other interferometric measurement systems.
- Open Access
GenéLive! Generating Rhythm Actions in Love Live!
This article presents our generative model for rhythm action games together with applications in business operations. Rhythm action games are video games in which the player is challenged to issue commands at the right timings during a music session. The timings are rendered in the chart, which consists of visual symbols, called notes, flying through the screen. We introduce our deep generative model, Gen\'eLive!, which outperforms the state-of-the-art model by taking into account musical structures through beats and temporal scales. Thanks to its favorable performance, Gen\'eLive! was put into operation at KLab Inc., a Japan-based video game developer, and reduced the business cost of chart generation by as much as half. The application target included the phenomenal "Love Live!," which has more than 10 million users across Asia and beyond, and is one of the few rhythm action franchises that has led the online era of the genre. In this article, we evaluate the generative performance of Gen\'eLive! using production datasets at KLab as well as open datasets for reproducibility, while the model continues to operate in their business. Our code and the model, tuned and trained using a supercomputer, are publicly available.
Open AccessGenéLive! Generating Rhythm Actions in Love Live!
This article presents our generative model for rhythm action games together with applications in business operations. Rhythm action games are video games in which the player is challenged to issue commands at the right timings during a music session. The timings are rendered in the chart, which consists of visual symbols, called notes, flying through the screen. We introduce our deep generative model, Gen\'eLive!, which outperforms the state-of-the-art model by taking into account musical structures through beats and temporal scales. Thanks to its favorable performance, Gen\'eLive! was put into operation at KLab Inc., a Japan-based video game developer, and reduced the business cost of chart generation by as much as half. The application target included the phenomenal "Love Live!," which has more than 10 million users across Asia and beyond, and is one of the few rhythm action franchises that has led the online era of the genre. In this article, we evaluate the generative performance of Gen\'eLive! using production datasets at KLab as well as open datasets for reproducibility, while the model continues to operate in their business. Our code and the model, tuned and trained using a supercomputer, are publicly available.
- Open Access
- 10 September 2022
- Published by Sumart Processing Society for Minerals, Environment and Energy in Journal of Smart Processing
- Vol. 11 (5) , 202-208
Visualization of Joint Behavior during Ultrasonic Flip Chip Bonding Using In-Situ Strain Monitoring Method
The purpose of this study is to clarify the joint behavior during ultrasonic flip chip bonding using the in-situ bonding monitoring method with the semiconductor strain sensor. Some characteristic tendencies related with the bonding strength ware extracted from the strain waves below the joint region. The higher the bonding strength was, the bigger reduction of the vertical compression strain during applying ultrasonic vibration was. Furthermore, the joint status was found to be correlated with the change in the amplitude and harmonic components of the horizontal strain with time. Especially, the difference in harmonic components of each wave was very useful as the future value of the machine learning algorithms for judging bonding qualities.
- 10 September 2022
- Published by Sumart Processing Society for Minerals, Environment and Energy in Journal of Smart Processing
- Vol. 11 (5) , 202-208
Open AccessVisualization of Joint Behavior during Ultrasonic Flip Chip Bonding Using In-Situ Strain Monitoring Method
The purpose of this study is to clarify the joint behavior during ultrasonic flip chip bonding using the in-situ bonding monitoring method with the semiconductor strain sensor. Some characteristic tendencies related with the bonding strength ware extracted from the strain waves below the joint region. The higher the bonding strength was, the bigger reduction of the vertical compression strain during applying ultrasonic vibration was. Furthermore, the joint status was found to be correlated with the change in the amplitude and harmonic components of the horizontal strain with time. Especially, the difference in harmonic components of each wave was very useful as the future value of the machine learning algorithms for judging bonding qualities.
- Open Access
- 20 July 2021
- Published by Springer Nature in The Visual Computer
- Vol. 37 (12) , 3039-3050
Visual analysis of geospatial multivariate data for investigating radioactive deposition processes
The Fukushima nuclear accident of 2011 raised awareness of the importance of radioactive deposition processes, especially for proposing aerosol measures against possible air pollution. However, identifying these types of processes is often difficult due to complicated terrains. This paper presents an application study for identifying radioactive deposition processes by taking advantage of visual interaction with topographic data. The idea is to visually investigate the correspondence of the spatial positions to the air dose rate along with relevant attributes. This is accomplished by composing scatterplots of pairwise attributes, onto which we project terrain areas to interactively find specific patterns of such attributes. We applied our approach to the analysis of air dose rate distribution data around the Fukushima nuclear plant after the accident. Our visualization technique clearly distinguished contamination areas derived from different deposition processes and thus is useful for elucidation of the deposition process.
- 20 July 2021
- Published by Springer Nature in The Visual Computer
- Vol. 37 (12) , 3039-3050
Open AccessVisual analysis of geospatial multivariate data for investigating radioactive deposition processes
The Fukushima nuclear accident of 2011 raised awareness of the importance of radioactive deposition processes, especially for proposing aerosol measures against possible air pollution. However, identifying these types of processes is often difficult due to complicated terrains. This paper presents an application study for identifying radioactive deposition processes by taking advantage of visual interaction with topographic data. The idea is to visually investigate the correspondence of the spatial positions to the air dose rate along with relevant attributes. This is accomplished by composing scatterplots of pairwise attributes, onto which we project terrain areas to interactively find specific patterns of such attributes. We applied our approach to the analysis of air dose rate distribution data around the Fukushima nuclear plant after the accident. Our visualization technique clearly distinguished contamination areas derived from different deposition processes and thus is useful for elucidation of the deposition process.