瀋　迅SHEN Xunシン　ジン

業績情報

氏名・連絡先

• 氏名

  シン　ジン, 瀋　迅, SHEN Xun

主たる所属・職名

• 工学研究院　知能情報システム工学部門, 准教授

学位

• 博士（機械工学）
  上智大学
  2018年03月31日

科学研究費助成事業

• 基盤研究（C）
  注視行動に基づく個別適合型高齢歩行者横断教育システムの開発
  自 2025年, 至 2025年
• 基盤研究（C）
  DNNを用いたフィードバック制御器の汎化性能向上技術の構築と実証的検証
  自 2025年, 至 2025年
• 若手研究
  個人特性への理解に基づくインタラクティブな高齢者運転教育システムの開発
  自 2024年, 至 2026年

論文

• Approximate Uncertain Program
  Shen, Xun; Zhuang, Jiancang; Zhang, Xingguo
  IEEE ACCESS
  IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
  Chance constrained program where one seeks to minimize an objective over decisions which satisfy randomly disturbed constraints with a given probability is computationally intractable. This paper proposes an approximate approach to address chance constrained program. Firstly, a single layer neural-network is used to approximate the function from decision domain to violation probability domain. The algorithm for updating parameters in single layer neural-network adopts sequential extreme learning machine. Based on the neural violation probability approximate model, a randomized algorithm is then proposed to approach the optimizer in the probabilistic feasible domain of decision. In the randomized algorithm, samples are extracted from decision domain uniformly at first. Then, violation probabilities of all samples are calculated according to neural violation probability approximate model. The ones with violation probability higher than the required level are discarded. The minimizer in the remained feasible decision samples is used to update sampling policy. The policy converges to the optimal feasible decision. Numerical simulations are implemented to validate the proposed method for non-convex problems comparing with scenario approach and parallel randomized algorithm. The results show that proposed method have improved performance.
  2019年, 研究論文（学術雑誌）, 共同, 7, 2169-3536, DOI(公開)(r-map), 182357, 182365
• Probabilistic reachable sets of stochastic nonlinear systems with contextual uncertainties
  Shen, Xun; Wang, Ye; Hashimoto, Kazumune; Wu, Yuhu; Gros, Sebastien
  AUTOMATICA
  PERGAMON-ELSEVIER SCIENCE LTD
  Validating and controlling safety-critical systems in uncertain environments necessitates probabilistic reachable sets of future state evolutions. The existing methods of computing probabilistic reachable sets normally assume that stochastic uncertainties are independent of system states, inputs, and other environment variables. However, this assumption falls short in many real-world applications, where the probability distribution governing uncertainties depends on these variables, referred to as contextual uncertainties. This paper addresses the challenge of computing probabilistic reachable sets of stochastic nonlinear states with contextual uncertainties by seeking minimum-volume polynomial sublevel sets with contextual chance constraints. The formulated problem cannot be solved by the existing sample-based approximation method since the existing methods do not consider conditional probability densities. To address this, we propose a consistent sample approximation of the original problem by leveraging conditional density estimation and resampling. The obtained approximate problem is a tractable optimization problem. Additionally, we prove the proposed sample-based approximation's almost uniform convergence, showing that it gives the optimal solution almost consistently with the original ones. Through a numerical example, we evaluate the effectiveness of the proposed method against existing approaches, highlighting its capability to significantly reduce the bias inherent in sample-based approximation without considering a conditional probability density. (c) 2025 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
  2025年06月, 研究論文（学術雑誌）, 共同, 176, 0005-1098, DOI(公開)(r-map)
• Optimal targeted marketing strategy in multiple market systems based on social network
  Yue, Mingda; Zhang, Jingyu; Wu, Yuhu; Shen, Xun
  IFAC JOURNAL OF SYSTEMS AND CONTROL
  ELSEVIER
  This paper investigates a duopoly marketing competition over multiple interconnected market systems (MSs). In each MS, consumers are divided into three groups: loyalists of Firm 1, loyalists of Firm 2, and undecided switchers. Firms employ targeted marketing strategies to influence consumer loyalty, leading to instantaneous shifts in the MS composition. Additionally, consumers across different MSs interact through a fixed social network, modeled as a directed graph, which drives continuous consensus-based opinion dynamics. We first establish that the consumer loyalty dynamics are well-posed over time. Then, we prove that the competition between the two firms always admits a unique Nash equilibrium. Furthermore, we analytically characterize the firms’ optimal advertising strategies at equilibrium by explicitly deriving the closed-form structure of their best responses. The results offer practical insights for managers on how to leverage social network interactions across MSs to optimize marketing resource allocation and competitive positioning.
  2025年12月, 研究論文（学術雑誌）, 共同, 35, DOI(公開)(r-map)
• Optimal Competitive Strategies: Pricing and Advertising in Dynamic Market Segments
  Yue, Mingda; Le, Shuting; Wu, Yuhu; Shen, Xun
  IEEE CONTROL SYSTEMS LETTERS
  IEEE
  This letter investigates a duopolistic market where firms compete through both pricing and advertising strategies. Due to customer attrition and the effects of advertising, each firm’s loyal customer segment evolves dynamically. This leads to a dynamic market segment competition in which firms aim to maximize profits through optimal strategies. Our analysis of the one-stage game reveals at most two Nash equilibria. Notably, even with symmetric initial positions, firms adopt asymmetric strategies in equilibrium: one employs greater advertising intensity and higher pricing to secure more profits. In the two-stage game, our analysis shows that firms optimally increase first-stage advertising to sustain larger loyal segments, thereby boosting subsequent profits. These findings have important implications for multi-stage competition.
  2025年10月17日, 研究論文（学術雑誌）, 共同, 9, 2475-1456, DOI(公開)(r-map), 2447, 2452
• Learning-Based Event-Triggered MPC With Gaussian Processes Under Terminal Constraints
  Hashimoto, Kazumune; Onoue, Yuga; Wachi, Akifumi; Shen, Xun
  IEEE TRANSACTIONS ON CYBERNETICS
  IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
  The event-triggered control strategy is capable of significantly reducing the number of control task executions while achieving desired control objectives, such as stability. In this article, we introduce a novel learning-based method for event-triggered model predictive control with initially unknown dynamics. The formulation of optimal control problems (OCPs) is based on predictive states derived from Gaussian process (GP) regression under terminal constraints. The event-triggered condition proposed in this article is derived from the recursive feasibility, so that the OCPs are solved only when an error between the predictive and the actual states exceeds a certain threshold. This article analyzes the convergence of the closed-loop system under the event-triggered condition, demonstrating that the system's state will enter the terminal set within a finite time, assuming small-enough uncertainty in the GP model. We validate this approach through a tracking control problem, illustrating its practical effectiveness.
  2025年04月, 研究論文（学術雑誌）, 共同, 55, 4, 2168-2267, DOI(公開)(r-map), 1512
• Risk-Regularization Optimization for Learning Wind Power Curve
  Zhang, Zhi; Ouyang, Tinghui; Shen, Xun
  IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
  IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
  Wind power prediction requires an accurate estimation of the wind power curve. However, anomalous data in the SCADA dataset deteriorates the performance of regression models. This paper proposes a novel risk-regularization optimization-based method that formulates a chance-constrained optimization problem for training an interval neural network. This method ensures that normal data is preserved with high probability while anomalous data is removed. This approach overcomes the limitations of existing anomaly detection techniques that rely on hyperparameter tuning and reference datasets. To address the computational challenges of solving chance-constrained programs, we introduce a sample-based risk-regularization method that enables efficient optimization. Experimental validation using real-world wind turbine data demonstrates the superiority of the proposed method. The results show that our approach improves anomaly detection and power curve regression performance over conventional methods. These findings confirm the effectiveness of our approach in enhancing wind power curve estimation.
  2025年09月01日, 研究論文（学術雑誌）, 共同, 2471-285X, DOI(公開)(r-map)
• Spatial-Temporal Optimal Pricing for Charging Stations: A Model-Driven Approach Based on Group Price Response Behavior of EVs
  Yang, Nan; Shen, Xun; Liang, Pengcheng; Ding, Li; Yan, Jing; Xing, Chao; Wang, Can; Zhang, Lei
  IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION
  IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
  To adapt to the dual carbon goals and effectively guide the charging of electric vehicles (EVs) while reducing issues such as long queueing times and underutilization of charging stations (CSs) due to unreasonable pricing by CS operators (CSOs), this article proposes an optimal spatial-temporal pricing strategy for CSs based on the group price response behavior of EVs. First, this study predicts the spatial-temporal distribution of EV load demand using trip chain and probability theory. Then, the Monte Carlo method is employed to simulate the spatial-temporal distribution of EV load demand and group charging behavior. Subsequently, an EV-CSO two-layer pricing demand response model is established, comprising an upper layer pricing model for CSOs and a lower layer charging decision model for the EV group. Finally, the model is solved to obtain the optimal pricing strategy. In addition, the decision behavior of EVs is simplified through node clustering, and the optimal spacing is obtained through the iterative search algorithm. The results show that compared to traditional pricing strategies, the proposed method improves the total profit of CSs and the average utilization rate of charging piles. Furthermore, the node clustering method significantly improves the computational efficiency of the pricing model, providing theoretical guidance for complex traffic network analysis of large-scale EVs.
  2024年12月, 研究論文（学術雑誌）, 共同, 10, 4, 2332-7782, DOI(公開)(r-map), 8869
• Integration of Dynamical Network Biomarkers, Control Theory and Drosophila Model Identifies Vasa/DDX4 as the Potential Therapeutic Targets for Metabolic Syndrome
  Akagi, Kazutaka; Jin, Ying-Jie; Koizumi, Keiichi; Oku, Makito; Ito, Kaisei; Shen, Xun; Imura, Jun-ichi; Aihara, Kazuyuki; Saito, Shigeru
  CELLS
  MDPI
  Metabolic syndrome (MetS) is a subclinical disease, resulting in increased risk of type 2 diabetes (T2D), cardiovascular diseases, cancer, and mortality. Dynamical network biomarkers (DNB) theory has been developed to provide early-warning signals of the disease state during a preclinical stage. To improve the efficiency of DNB analysis for the target genes discovery, the DNB intervention analysis based on the control theory has been proposed. However, its biological validation in a specific disease such as MetS remains unexplored. Herein, we identified eight candidate genes from adipose tissue of MetS model mice at the preclinical stage by the DNB intervention analysis. Using Drosophila, we conducted RNAi-mediated knockdown screening of these candidate genes and identified vasa (also known as DDX4), encoding a DEAD-box RNA helicase, as a fat metabolism-associated gene. Fat body-specific knockdown of vasa abrogated high-fat diet (HFD)-induced enhancement of starvation resistance through up-regulation of triglyceride lipase. We also confirmed that DDX4 expressing adipocytes are increased in HFD-fed mice and high BMI patients using the public datasets. These results prove the potential of the DNB intervention analysis to search the therapeutic targets for diseases at the preclinical stage.
  2025年03月12日, 研究論文（学術雑誌）, 共同, 14, 6, DOI(公開)(r-map)