Yusuke Uchiyama
Research Subjects
Yusuke Uchiyama has a total of 26 co-authors. Here are the top co-authors by number of shared publications:
Hidetoshi Konno
Department of Risk Engineering, Faculty of Information and Systems, University of Tsukuba, Tsukuba, Ibaraki 305-8573, Japan
6 shared publications- Kei Nakagawa
Osaka Metropolitan University, Osaka, Japan
5 shared publications Jun Nakano
Department of Information and Computer Science, Kanazawa Institute of Technology
2 shared publicationsKenji Kansaku
Department of Physiology, Dokkyo Medical University, Graduate School of Medicine
2 shared publicationsKyoko Aoyagi
Department of Neurosurgery, Chiba Cerebral and Cardiovascular Center
2 shared publications- Shota Yabui
Faculty of Science and Engineering, Tokyo City University, Tokyo, Japan
2 shared publications
Latest Publications
- Open Access
- 17 August 2026
- Published by Frontiers Media SA in Frontiers in Neuroscience
- Vol. 20, 1864936
A knowledge-based Bayesian network reveals shared prediction errors with physicians in Gamma Knife radiosurgery
Although Artificial Intelligence (AI) has shown strong predictive performance in medicine, its role in providing cognitive support remains unclear. We developed a knowledge-based expert system using a Bayesian Network (BN) to explicitly model an expert neurosurgeon's clinical reasoning for predicting outcomes of Gamma Knife radiosurgery. The BN structure was constructed from elicited expert knowledge and encoded as a directed acyclic (DAG) graph that represents clinically relevant variables and their interrelationships. Model parameters were estimated using a clinical dataset to predict patient overall survival (OS). In parallel, physicians independently estimated outcomes for the same patients, enabling a direct comparison of predictive performance and error characteristics. The BN captured clinically plausible reasoning patterns and produced less skewed prediction errors than physicians, suggesting partial mitigation of individual cognitive bias. However, both the BN and physicians exhibited a heavy-tailed error distribution, systematically failing to predict patients with unexpectedly long-term survival. This convergence of error patterns indicates that both human and model-based predictions are constrained by unobserved or unmodeled factors, limiting their ability to capture exceptional outcomes. These findings highlight a limitation of predictive paradigms based on current expert knowledge and suggest the need for iterative refinement of knowledge-based models. Incorporating additional expert knowledge and updating the underlying structure may help identify previously unmodeled relationships and reduce unexplained variance. Taken together, our results position knowledge-based BNs not only as interpretable predictive tools, but also as evolving frameworks for integrating clinical knowledge and probing the shared limits of human-AI decision-making, highlighting their potential as computational cognitive neuroprosthetic systems that aim to augment physicians' clinical reasoning under uncertainty.
- 17 August 2026
- Published by Frontiers Media SA in Frontiers in Neuroscience
- Vol. 20, 1864936
Open AccessA knowledge-based Bayesian network reveals shared prediction errors with physicians in Gamma Knife radiosurgery
Although Artificial Intelligence (AI) has shown strong predictive performance in medicine, its role in providing cognitive support remains unclear. We developed a knowledge-based expert system using a Bayesian Network (BN) to explicitly model an expert neurosurgeon's clinical reasoning for predicting outcomes of Gamma Knife radiosurgery. The BN structure was constructed from elicited expert knowledge and encoded as a directed acyclic (DAG) graph that represents clinically relevant variables and their interrelationships. Model parameters were estimated using a clinical dataset to predict patient overall survival (OS). In parallel, physicians independently estimated outcomes for the same patients, enabling a direct comparison of predictive performance and error characteristics. The BN captured clinically plausible reasoning patterns and produced less skewed prediction errors than physicians, suggesting partial mitigation of individual cognitive bias. However, both the BN and physicians exhibited a heavy-tailed error distribution, systematically failing to predict patients with unexpectedly long-term survival. This convergence of error patterns indicates that both human and model-based predictions are constrained by unobserved or unmodeled factors, limiting their ability to capture exceptional outcomes. These findings highlight a limitation of predictive paradigms based on current expert knowledge and suggest the need for iterative refinement of knowledge-based models. Incorporating additional expert knowledge and updating the underlying structure may help identify previously unmodeled relationships and reduce unexplained variance. Taken together, our results position knowledge-based BNs not only as interpretable predictive tools, but also as evolving frameworks for integrating clinical knowledge and probing the shared limits of human-AI decision-making, highlighting their potential as computational cognitive neuroprosthetic systems that aim to augment physicians' clinical reasoning under uncertainty.
- Open Access
Development of a Control System for a Hydraulic Injection Molding Machine Using an AFC Controller and Utilization of Learning Parameters
Maintaining stable molding quality in hydraulic injection molding machines is difficult because the internal state of molten resin cannot be directly observed and varies with material properties and operating conditions. This difficulty is intensified by variations in hydraulic characteristics caused by oil temperature changes. This study proposes an adaptive feedforward control (AFC) framework that improves injection velocity tracking while utilizing AFC learning parameters as indicators of resin state. AFC is implemented as a multi-frequency feedforward controller whose parameters are updated through repetitive injection cycles. To overcome the limited learning duration within a single injection shot, a shot-to-shot compensation mechanism accumulates and transfers learning results across consecutive shots. Experiments are conducted on a hydraulic injection molding machine using polypropylene materials with different viscosities. The results show that the converged AFC learning parameters vary systematically with material changes and correspond to differences in molded product appearance. Furthermore, by adjusting the cylinder temperature of another material, the AFC parameters converge to values close to those of a reference material, resulting in similar molded products. These findings demonstrate that AFC learning parameters reflect variations in resin state and can serve as practical state indicators for aligning molding conditions.
Open AccessDevelopment of a Control System for a Hydraulic Injection Molding Machine Using an AFC Controller and Utilization of Learning Parameters
Maintaining stable molding quality in hydraulic injection molding machines is difficult because the internal state of molten resin cannot be directly observed and varies with material properties and operating conditions. This difficulty is intensified by variations in hydraulic characteristics caused by oil temperature changes. This study proposes an adaptive feedforward control (AFC) framework that improves injection velocity tracking while utilizing AFC learning parameters as indicators of resin state. AFC is implemented as a multi-frequency feedforward controller whose parameters are updated through repetitive injection cycles. To overcome the limited learning duration within a single injection shot, a shot-to-shot compensation mechanism accumulates and transfers learning results across consecutive shots. Experiments are conducted on a hydraulic injection molding machine using polypropylene materials with different viscosities. The results show that the converged AFC learning parameters vary systematically with material changes and correspond to differences in molded product appearance. Furthermore, by adjusting the cylinder temperature of another material, the AFC parameters converge to values close to those of a reference material, resulting in similar molded products. These findings demonstrate that AFC learning parameters reflect variations in resin state and can serve as practical state indicators for aligning molding conditions.
- Open Access
- 28 October 2025
- Published by The Institute of Systems, Control and Information Engineers in Proceedings of the ISCIE International Symposium on Stochastic Systems Theory and its Applications
- Vol. 2025, 30-35
Modeling of Electric Servo Injection Molding Machine Based on Gray Box Model
This study focuses on the development of a simulation model for an electric servo injection molding machine. Plastic products have become indispensable in a wide variety of fields due to their light weight, durability, formability, and insulation properties. Injection molding, an important process in its production, enables highly efficient mass production and high-precision shape molding. However, it faces challenges such as dependence on skilled engineers, variation in molding quality due to trial and error in the prototype stage, and waste of resources. To address these issues, this study employed a gray-box modeling approach that utilizes actual equipment data. In this study, we focused on injection equipment and built a base model to reproduce the molding process using Simscape (MATLAB). Parameters were then adjusted through Bayesian optimization to improve simulation accuracy. Initial results confirmed the validity of the model, and subsequent adjustments have significantly improved the correlation between simulation output and actual equipment data. The results of this research show the potential to contribute to efficient optimization of molding conditions, reduction of waste in the prototype stage, and improvement of operational efficiency.
- 28 October 2025
- Published by The Institute of Systems, Control and Information Engineers in Proceedings of the ISCIE International Symposium on Stochastic Systems Theory and its Applications
- Vol. 2025, 30-35
Open AccessModeling of Electric Servo Injection Molding Machine Based on Gray Box Model
This study focuses on the development of a simulation model for an electric servo injection molding machine. Plastic products have become indispensable in a wide variety of fields due to their light weight, durability, formability, and insulation properties. Injection molding, an important process in its production, enables highly efficient mass production and high-precision shape molding. However, it faces challenges such as dependence on skilled engineers, variation in molding quality due to trial and error in the prototype stage, and waste of resources. To address these issues, this study employed a gray-box modeling approach that utilizes actual equipment data. In this study, we focused on injection equipment and built a base model to reproduce the molding process using Simscape (MATLAB). Parameters were then adjusted through Bayesian optimization to improve simulation accuracy. Initial results confirmed the validity of the model, and subsequent adjustments have significantly improved the correlation between simulation output and actual equipment data. The results of this research show the potential to contribute to efficient optimization of molding conditions, reduction of waste in the prototype stage, and improvement of operational efficiency.
- Open Access
- 18 November 2022
- Published by Japanese Association of Rehabilitation Medicine in The Japanese Journal of Rehabilitation Medicine
- Vol. 59 (11) , 1100-1105
Applications of Bayesian Network for Medical and Clinical Data
本稿では,解釈性と説明性を有するAIモデルであるベイジアンネットワークの概要と医学分野へのいくつかの応用事例について紹介する.合わせて,ベイジアンネットワークの実データへの応用例の1つとして,筆者が所属する研究グループの最新の結果である,定位放射線治療後の転帰予測について紹介する.
- 18 November 2022
- Published by Japanese Association of Rehabilitation Medicine in The Japanese Journal of Rehabilitation Medicine
- Vol. 59 (11) , 1100-1105
Open AccessApplications of Bayesian Network for Medical and Clinical Data
本稿では,解釈性と説明性を有するAIモデルであるベイジアンネットワークの概要と医学分野へのいくつかの応用事例について紹介する.合わせて,ベイジアンネットワークの実データへの応用例の1つとして,筆者が所属する研究グループの最新の結果である,定位放射線治療後の転帰予測について紹介する.
- Open Access
Entropy Based Student’s t-Process Dynamical Model
Volatility, which represents the magnitude of fluctuating asset prices or returns, is used in the problems of finance to design optimal asset allocations and to calculate the price of derivatives. Since volatility is unobservable, it is identified and estimated by latent variable models known as volatility fluctuation models. Almost all conventional volatility fluctuation models are linear time-series models and thus are difficult to capture nonlinear and/or non-Gaussian properties of volatility dynamics. In this study, we propose an entropy based Student’s t-process Dynamical model (ETPDM) as a volatility fluctuation model combined with both nonlinear dynamics and non-Gaussian noise. The ETPDM estimates its latent variables and intrinsic parameters by a robust particle filtering based on a generalized H-theorem for a relative entropy. To test the performance of the ETPDM, we implement numerical experiments for financial time-series and confirm the robustness for a small number of particles by comparing with the conventional particle filtering.
Open AccessEntropy Based Student’s t-Process Dynamical Model
Volatility, which represents the magnitude of fluctuating asset prices or returns, is used in the problems of finance to design optimal asset allocations and to calculate the price of derivatives. Since volatility is unobservable, it is identified and estimated by latent variable models known as volatility fluctuation models. Almost all conventional volatility fluctuation models are linear time-series models and thus are difficult to capture nonlinear and/or non-Gaussian properties of volatility dynamics. In this study, we propose an entropy based Student’s t-process Dynamical model (ETPDM) as a volatility fluctuation model combined with both nonlinear dynamics and non-Gaussian noise. The ETPDM estimates its latent variables and intrinsic parameters by a robust particle filtering based on a generalized H-theorem for a relative entropy. To test the performance of the ETPDM, we implement numerical experiments for financial time-series and confirm the robustness for a small number of particles by comparing with the conventional particle filtering.
- Open Access
- 7 November 2020
- Published by MDPI AG in Mathematics
- Vol. 8 (11) , 1990
GO-GJRSK Model with Application to Higher Order Risk-Based Portfolio
There are three distinguishing features in the financial time series, such as stock prices, are as follows: (1) Non-normality, (2) serial correlation, and (3) leverage effect. All three points need to be taken into account to model the financial time series. However, multivariate financial time series modeling involves a large number of stocks, with many parameters to be estimated. Therefore, there are few examples of multivariate financial time series modeling that explicitly deal with higher-order moments. Furthermore, there is no multivariate financial time series model that takes all three characteristics above into account. In this study, we propose the generalized orthogonal (GO)-Glosten, Jagannathan, and Runkle GARCH (GJR) model which extends the GO-generalized autoregressive conditional heteroscedasticity (GARCH) model and incorporates the three features of the financial time series. We confirm the effectiveness of the proposed model by comparing the performance of risk-based portfolios with higher-order moments. The results show that the performance with our proposed model is superior to that with baseline methods, and indicate that estimation methods are important in risk-based portfolios with higher moments.
- 7 November 2020
- Published by MDPI AG in Mathematics
- Vol. 8 (11) , 1990
Open AccessGO-GJRSK Model with Application to Higher Order Risk-Based Portfolio
There are three distinguishing features in the financial time series, such as stock prices, are as follows: (1) Non-normality, (2) serial correlation, and (3) leverage effect. All three points need to be taken into account to model the financial time series. However, multivariate financial time series modeling involves a large number of stocks, with many parameters to be estimated. Therefore, there are few examples of multivariate financial time series modeling that explicitly deal with higher-order moments. Furthermore, there is no multivariate financial time series model that takes all three characteristics above into account. In this study, we propose the generalized orthogonal (GO)-Glosten, Jagannathan, and Runkle GARCH (GJR) model which extends the GO-generalized autoregressive conditional heteroscedasticity (GARCH) model and incorporates the three features of the financial time series. We confirm the effectiveness of the proposed model by comparing the performance of risk-based portfolios with higher-order moments. The results show that the performance with our proposed model is superior to that with baseline methods, and indicate that estimation methods are important in risk-based portfolios with higher moments.
- Open Access
- 19 March 2020
- Published by MDPI AG in Mathematics
- Vol. 8 (3) , 449
TPLVM: Portfolio Construction by Student’s t-Process Latent Variable Model
Optimal asset allocation is a key topic in modern finance theory. To realize the optimal asset allocation on investor’s risk aversion, various portfolio construction methods have been proposed. Recently, the applications of machine learning are rapidly growing in the area of finance. In this article, we propose the Student’s t-process latent variable model (TPLVM) to describe non-Gaussian fluctuations of financial timeseries by lower dimensional latent variables. Subsequently, we apply the TPLVM to portfolio construction as an alternative of existing nonlinear factor models. To test the performance of the proposed method, we construct minimum-variance portfolios of global stock market indices based on the TPLVM or Gaussian process latent variable model. By comparing these portfolios, we confirm the proposed portfolio outperforms that of the existing Gaussian process latent variable model.
- 19 March 2020
- Published by MDPI AG in Mathematics
- Vol. 8 (3) , 449
Open AccessTPLVM: Portfolio Construction by Student’s t-Process Latent Variable Model
Optimal asset allocation is a key topic in modern finance theory. To realize the optimal asset allocation on investor’s risk aversion, various portfolio construction methods have been proposed. Recently, the applications of machine learning are rapidly growing in the area of finance. In this article, we propose the Student’s t-process latent variable model (TPLVM) to describe non-Gaussian fluctuations of financial timeseries by lower dimensional latent variables. Subsequently, we apply the TPLVM to portfolio construction as an alternative of existing nonlinear factor models. To test the performance of the proposed method, we construct minimum-variance portfolios of global stock market indices based on the TPLVM or Gaussian process latent variable model. By comparing these portfolios, we confirm the proposed portfolio outperforms that of the existing Gaussian process latent variable model.
- Open Access
TPLVM: Portfolio Construction by Student's $t$-process Latent Variable Model
Optimal asset allocation is a key topic in modern finance theory. To realize the optimal asset allocation on investor's risk aversion, various portfolio construction methods have been proposed. Recently, the applications of machine learning are rapidly growing in the area of finance. In this article, we propose the Student's $t$-process latent variable model (TPLVM) to describe non-Gaussian fluctuations of financial timeseries by lower dimensional latent variables. Subsequently, we apply the TPLVM to minimum-variance portfolio as an alternative of existing nonlinear factor models. To test the performance of the proposed portfolio, we construct minimum-variance portfolios of global stock market indices based on the TPLVM or Gaussian process latent variable model. By comparing these portfolios, we confirm the proposed portfolio outperforms that of the existing Gaussian process latent variable model.
Open AccessTPLVM: Portfolio Construction by Student's $t$-process Latent Variable Model
Optimal asset allocation is a key topic in modern finance theory. To realize the optimal asset allocation on investor's risk aversion, various portfolio construction methods have been proposed. Recently, the applications of machine learning are rapidly growing in the area of finance. In this article, we propose the Student's $t$-process latent variable model (TPLVM) to describe non-Gaussian fluctuations of financial timeseries by lower dimensional latent variables. Subsequently, we apply the TPLVM to minimum-variance portfolio as an alternative of existing nonlinear factor models. To test the performance of the proposed portfolio, we construct minimum-variance portfolios of global stock market indices based on the TPLVM or Gaussian process latent variable model. By comparing these portfolios, we confirm the proposed portfolio outperforms that of the existing Gaussian process latent variable model.
- Open Access
Complex Valued Risk Diversification
Risk diversification is one of the dominant concerns for portfolio managers. Various portfolio constructions have been proposed to minimize the risk of the portfolio under some constraints, including expected returns. We propose a portfolio construction method that incorporates the complex valued principal component analysis into the risk diversification portfolio construction. The proposed method was verified to outperform the conventional risk parity and risk diversification portfolio constructions.
Open AccessComplex Valued Risk Diversification
Risk diversification is one of the dominant concerns for portfolio managers. Various portfolio constructions have been proposed to minimize the risk of the portfolio under some constraints, including expected returns. We propose a portfolio construction method that incorporates the complex valued principal component analysis into the risk diversification portfolio construction. The proposed method was verified to outperform the conventional risk parity and risk diversification portfolio constructions.
- Open Access
Variable coefficient complex Ginzburg-Landau equation
The complex Ginzburg-Landau equation (CGLE) is a general model of spatially extended nonequilibrium systems. In this paper, an analytical method for a variable coefficient CGLE is presented to obtain exact solutions. Variable transformations for space and time variables with coefficient functions yield an imaginary time advection equation related to a complex valued characteristic curve. The variable coefficient CGLE is transformed into the nonlinear Schr{"\o}dinger equation (NLSE) on the complex valued characteristic curve. This result indicates that the analytical solutions of the NLSE generate that of the variable coefficient CGLE.
Open AccessVariable coefficient complex Ginzburg-Landau equation
The complex Ginzburg-Landau equation (CGLE) is a general model of spatially extended nonequilibrium systems. In this paper, an analytical method for a variable coefficient CGLE is presented to obtain exact solutions. Variable transformations for space and time variables with coefficient functions yield an imaginary time advection equation related to a complex valued characteristic curve. The variable coefficient CGLE is transformed into the nonlinear Schr{"\o}dinger equation (NLSE) on the complex valued characteristic curve. This result indicates that the analytical solutions of the NLSE generate that of the variable coefficient CGLE.