Pub Date : 2025-09-10DOI: 10.1109/TSMC.2025.3606507
Xiujuan Ma;Xinwang Liu;Zaiwu Gong;Fang Liu
Individual interactions and conformity play a crucial role in shaping group opinions and influencing the decision-making process. This article introduces a novel opinion evolution model to simulate the impact of conformity on group opinion formation, focusing on weight allocation, relationship propagation, and evolution analysis. In the weight allocation phase, individual weights are evaluated using network structure and the PageRank algorithm. For relationship propagation, indirect trust relationships are computed via a weighted average method, leading to a more precise social network model. In the evolution analysis, an improved Hegselmann–Krause (HK) model demonstrates evolutionary stability. Two types of conformity behavior are simulated: active conformity behavior within clusters and passive conformity behavior via opinion leaders across clusters. Experimental studies on public health policy validate the effectiveness of the proposed model. The results show that this model more accurately captures the complex behavioral patterns of individuals in large-scale social networks and exhibits strong evolutionary stability.
{"title":"Modeling Opinion Evolution and Conformity Behavior in Large-Scale Social Network Group Decision-Making","authors":"Xiujuan Ma;Xinwang Liu;Zaiwu Gong;Fang Liu","doi":"10.1109/TSMC.2025.3606507","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3606507","url":null,"abstract":"Individual interactions and conformity play a crucial role in shaping group opinions and influencing the decision-making process. This article introduces a novel opinion evolution model to simulate the impact of conformity on group opinion formation, focusing on weight allocation, relationship propagation, and evolution analysis. In the weight allocation phase, individual weights are evaluated using network structure and the PageRank algorithm. For relationship propagation, indirect trust relationships are computed via a weighted average method, leading to a more precise social network model. In the evolution analysis, an improved Hegselmann–Krause (HK) model demonstrates evolutionary stability. Two types of conformity behavior are simulated: active conformity behavior within clusters and passive conformity behavior via opinion leaders across clusters. Experimental studies on public health policy validate the effectiveness of the proposed model. The results show that this model more accurately captures the complex behavioral patterns of individuals in large-scale social networks and exhibits strong evolutionary stability.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 11","pages":"8558-8571"},"PeriodicalIF":8.7,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145335309","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-10DOI: 10.1109/TSMC.2025.3606222
Hai-Feng Zhang;Kun-Peng Liu;Huan Wang;Chuang Ma
Network reconstruction, which involves inferring a network’s topology from observational data, is a critical challenge in network science. Because observational data are often limited, prior knowledge is often employed to enhance the accuracy of network reconstruction, including constraints related to sparsity, symmetry, or network dynamics. In many cases, we may possess prior knowledge regarding the number of types of edges within networks, yet the specific solutions of these types of edges remain unknown. For instance, while it is evident that two types of edges (i.e., positive and negative) exist in signed networks, the possible solution for each edge type is frequently unclear, rendering existing methods, such as the signal Lasso approach, ineffective. In this work, to effectively leverage this readily available prior knowledge, we propose a novel network reconstruction framework based on the Gaussian mixture model (GMM), which integrates Bayesian models and employs the GMM to model the distribution of unknown edges as prior probabilities. The method is effective for both unsigned and signed networks and achieves high accuracy even with limited prior information, without requiring specific solutions for each edge type, particularly in cases of network sparsity or noisy data.
{"title":"Reconstructing Network Structures Using Gaussian Mixture Model: From Unsigned to Signed Networks","authors":"Hai-Feng Zhang;Kun-Peng Liu;Huan Wang;Chuang Ma","doi":"10.1109/TSMC.2025.3606222","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3606222","url":null,"abstract":"Network reconstruction, which involves inferring a network’s topology from observational data, is a critical challenge in network science. Because observational data are often limited, prior knowledge is often employed to enhance the accuracy of network reconstruction, including constraints related to sparsity, symmetry, or network dynamics. In many cases, we may possess prior knowledge regarding the number of types of edges within networks, yet the specific solutions of these types of edges remain unknown. For instance, while it is evident that two types of edges (i.e., positive and negative) exist in signed networks, the possible solution for each edge type is frequently unclear, rendering existing methods, such as the signal Lasso approach, ineffective. In this work, to effectively leverage this readily available prior knowledge, we propose a novel network reconstruction framework based on the Gaussian mixture model (GMM), which integrates Bayesian models and employs the GMM to model the distribution of unknown edges as prior probabilities. The method is effective for both unsigned and signed networks and achieves high accuracy even with limited prior information, without requiring specific solutions for each edge type, particularly in cases of network sparsity or noisy data.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 11","pages":"8544-8557"},"PeriodicalIF":8.7,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145335293","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In group decision-making (GDM), the opinions of decision-makers (DMs) are prone to having controversies and conflicts. Identifying individual personality traits can better predict the individual adjustment behavior of DMs in GDM and, therefore, construct the corresponding feedback strategies to guide opinion interaction and consensus building. To do that, large language models (LLMs) are utilized to analyze the individual Big Five personality traits revealed by online text information. Then, a conflict quadrant diagram (CQD) is developed to explore the conflict resolution behaviors manifested by DMs as influenced by their personality traits. Subsequently, a series of interaction rules corresponding to diverse conflict behaviors within the CQD are constructed, and then a personality traits-driven feedback model is proposed to generate personalized recommendation advice for group consensus interaction, with the overarching aim of effectively enhancing the level of group consensus. Finally, a simulation experiment on LLM-based agents is conducted to verify the opinion convergence process, and some sensitivity and comparative analyses are also provided. Overall, this article contributes to the innovative application of LLMs in solving GDM problems by prompt engineering to generate outputs and validate models and carries out in-depth explorations on integrating individual personality traits into the group consensus-building process.
{"title":"A Personality Traits-Driven Conflict Quadrant Diagram by Large Language Models for Personalized Feedback in Group Decision-Making","authors":"Tiantian Gai;Jian Wu;Francisco Chiclana;Mi Zhou;Witold Pedrycz","doi":"10.1109/TSMC.2025.3605404","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3605404","url":null,"abstract":"In group decision-making (GDM), the opinions of decision-makers (DMs) are prone to having controversies and conflicts. Identifying individual personality traits can better predict the individual adjustment behavior of DMs in GDM and, therefore, construct the corresponding feedback strategies to guide opinion interaction and consensus building. To do that, large language models (LLMs) are utilized to analyze the individual Big Five personality traits revealed by online text information. Then, a conflict quadrant diagram (CQD) is developed to explore the conflict resolution behaviors manifested by DMs as influenced by their personality traits. Subsequently, a series of interaction rules corresponding to diverse conflict behaviors within the CQD are constructed, and then a personality traits-driven feedback model is proposed to generate personalized recommendation advice for group consensus interaction, with the overarching aim of effectively enhancing the level of group consensus. Finally, a simulation experiment on LLM-based agents is conducted to verify the opinion convergence process, and some sensitivity and comparative analyses are also provided. Overall, this article contributes to the innovative application of LLMs in solving GDM problems by prompt engineering to generate outputs and validate models and carries out in-depth explorations on integrating individual personality traits into the group consensus-building process.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 11","pages":"8506-8518"},"PeriodicalIF":8.7,"publicationDate":"2025-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145335325","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-05DOI: 10.1109/TSMC.2025.3605349
Lifei Dai;Changzhu Zhang;Hao Zhang;Yuxiong Ji;Huaicheng Yan
This article introduces a sustainable human-guided reinforcement learning (RL) framework to address the challenge of learning performance degradation when the human guidance is suspended. First, a compensation reward based on the historical similarity between the RL agent and human guidance history is designed to ensure the continued influence of human guidance. To avoid cumulative errors in value function approximation caused by fitting the new reward, including the compensation reward, a novel RL paradigm is proposed, which bypasses value function fitting and directly optimizes the policy using historical similarity. This paradigm develops a new historical similarity-based learning objective for RL to leverage human guidance more efficiently and achieve alignment with human behavior. Furthermore, the proposed paradigm enables the fine-tuning of the RL agent to address the long-tail problem. Experimental results demonstrate the advantages of the proposed method in terms of sustainable guidance and optimal performance in the autonomous driving, achieving a 15% increase in optimal performance compared with existing state-of-the-art (SOTA) methods.
{"title":"Sustainable Reinforcement Learning for Autonomous Driving Under Postsuspension of Human Guidance","authors":"Lifei Dai;Changzhu Zhang;Hao Zhang;Yuxiong Ji;Huaicheng Yan","doi":"10.1109/TSMC.2025.3605349","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3605349","url":null,"abstract":"This article introduces a sustainable human-guided reinforcement learning (RL) framework to address the challenge of learning performance degradation when the human guidance is suspended. First, a compensation reward based on the historical similarity between the RL agent and human guidance history is designed to ensure the continued influence of human guidance. To avoid cumulative errors in value function approximation caused by fitting the new reward, including the compensation reward, a novel RL paradigm is proposed, which bypasses value function fitting and directly optimizes the policy using historical similarity. This paradigm develops a new historical similarity-based learning objective for RL to leverage human guidance more efficiently and achieve alignment with human behavior. Furthermore, the proposed paradigm enables the fine-tuning of the RL agent to address the long-tail problem. Experimental results demonstrate the advantages of the proposed method in terms of sustainable guidance and optimal performance in the autonomous driving, achieving a 15% increase in optimal performance compared with existing state-of-the-art (SOTA) methods.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 11","pages":"8519-8530"},"PeriodicalIF":8.7,"publicationDate":"2025-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145335283","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-28DOI: 10.1109/TSMC.2025.3599146
Hang Su;Weihai Zhang
It is a common control issue that the input signal of the system is quantized in the controller-to-actuator channel via the communication network, but few results are available in considering adaptive tracking control problem for nonstrict-feedback nonlinear system with both state and input quantization. The control problem is figured out in our article by developing an adaptive neural network control method for nonstrict-feedback nonlinear system with quantized input and states. In addition to overcoming the difficulty that the virtual control signal cannot be defined by quantized states in backstepping-based design approach, our work also surmounts the influence of the coexistence of nonstrict-feedback structure and state discontinuity resulted from quantization, and gives the construction of adaptive law for the weight vector of approximation system based on neural network. Elaborate simulation examples are depicted to verify the effectiveness of our depicted quantized control algorithm.
{"title":"Adaptive Neural Network Tracking Control for Nonstrict-Feedback Nonlinear Systems With States and Inputs Quantization","authors":"Hang Su;Weihai Zhang","doi":"10.1109/TSMC.2025.3599146","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3599146","url":null,"abstract":"It is a common control issue that the input signal of the system is quantized in the controller-to-actuator channel via the communication network, but few results are available in considering adaptive tracking control problem for nonstrict-feedback nonlinear system with both state and input quantization. The control problem is figured out in our article by developing an adaptive neural network control method for nonstrict-feedback nonlinear system with quantized input and states. In addition to overcoming the difficulty that the virtual control signal cannot be defined by quantized states in backstepping-based design approach, our work also surmounts the influence of the coexistence of nonstrict-feedback structure and state discontinuity resulted from quantization, and gives the construction of adaptive law for the weight vector of approximation system based on neural network. Elaborate simulation examples are depicted to verify the effectiveness of our depicted quantized control algorithm.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 10","pages":"7591-7602"},"PeriodicalIF":8.7,"publicationDate":"2025-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145090121","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Evolutionary many-task optimization (EMaTO) endeavors to solve more than three optimization tasks simultaneously by leveraging similarities among tasks. While existing algorithms have shown promising results, they face significant challenges in low-similarity scenarios. First, existing transfer techniques, which rely on population location and distribution, become ineffective. Second, the difficulty of selecting appropriate knowledge increases significantly. To address these challenges, we introduce a new concept: distribution direction knowledge, i.e., the evolutionary direction (ED) of elite solutions. It enables the target task to learn the search experience of source tasks with similar evolutionary trends. To utilize this knowledge effectively, an EMaTO algorithm with distribution direction-assist two-stage knowledge transfer (DTSKT) is proposed. First, an ED-based multisource selection strategy is proposed to obtain appropriate knowledge in different circumstances. Second, we design a two-stage knowledge transfer (TSKT) strategy to search promising regions, consisting of exploration-oriented and exploitation-oriented knowledge transfer. In addition, to directly obtain distribution direction knowledge, the estimation of distribution algorithm is applied as the basic optimizer, explicitly revealing the ED of populations by employing probability distributions. Afterward, to validate the ability of DTSKT to handle tasks with different similarities, we utilize a test problem generator to create a more challenging many-task benchmark suite, named STOP. The results on the WCCI20 and STOP benchmark suites, along with a real-world application, demonstrate that DTSKT generally outperforms seven state-of-the-art algorithms.
{"title":"Distribution Direction-Assisted Two-Stage Knowledge Transfer for Many-Task Optimization","authors":"Tingyu Zhang;Xinyi Wu;Yanchi Li;Wenyin Gong;Hu Qin","doi":"10.1109/TSMC.2025.3598800","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3598800","url":null,"abstract":"Evolutionary many-task optimization (EMaTO) endeavors to solve more than three optimization tasks simultaneously by leveraging similarities among tasks. While existing algorithms have shown promising results, they face significant challenges in low-similarity scenarios. First, existing transfer techniques, which rely on population location and distribution, become ineffective. Second, the difficulty of selecting appropriate knowledge increases significantly. To address these challenges, we introduce a new concept: distribution direction knowledge, i.e., the evolutionary direction (ED) of elite solutions. It enables the target task to learn the search experience of source tasks with similar evolutionary trends. To utilize this knowledge effectively, an EMaTO algorithm with distribution direction-assist two-stage knowledge transfer (DTSKT) is proposed. First, an ED-based multisource selection strategy is proposed to obtain appropriate knowledge in different circumstances. Second, we design a two-stage knowledge transfer (TSKT) strategy to search promising regions, consisting of exploration-oriented and exploitation-oriented knowledge transfer. In addition, to directly obtain distribution direction knowledge, the estimation of distribution algorithm is applied as the basic optimizer, explicitly revealing the ED of populations by employing probability distributions. Afterward, to validate the ability of DTSKT to handle tasks with different similarities, we utilize a test problem generator to create a more challenging many-task benchmark suite, named STOP. The results on the WCCI20 and STOP benchmark suites, along with a real-world application, demonstrate that DTSKT generally outperforms seven state-of-the-art algorithms.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 10","pages":"7551-7565"},"PeriodicalIF":8.7,"publicationDate":"2025-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145090119","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-26DOI: 10.1109/TSMC.2025.3599882
Changzhong Wang;Yang Zhang;Shuang An;Weiping Ding
Fuzzy rough set theory is an important approach for analyzing data uncertainty. However, the model lacks adaptive learning capabilities and cannot fit labeled data effectively in classification tasks. This study aims to introduce an adaptive learning mechanism into fuzzy rough set theory to enhance its data-fitting capability. To this end, this study seamlessly integrates fuzzy rough set theory with neural networks and proposes a novel fuzzy rough neural network model. This model adaptively learns fuzzy similarity relations in rough set models using the backpropagation algorithm. The proposed network model comprises five layers: the input, membership, fuzzy lower approximation, fully connected, and output layers. The fuzzy similarity relations between the input samples and training samples are computed in the membership layer. These relations are utilized in the fuzzy lower approximation layer to describe the degree to which the samples belong to different classes. The fuzzy rough lower approximations of the input samples are finally fused in the fully connected layer using feature weight coefficients. In the backpropagation stage, the gradient of the objective function is used to correct the fuzzy similarity relations and feature weight coefficients. This study theoretically proved that the proposed fuzzy rough network has a generalized function approximation property and can approximate any decision function. Experimental analysis showed that the proposed method is effective and performs better than most of the existing state-of-the-art algorithms.
{"title":"An Adaptive Fuzzy Rough Neural Network and Its Application in Classification","authors":"Changzhong Wang;Yang Zhang;Shuang An;Weiping Ding","doi":"10.1109/TSMC.2025.3599882","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3599882","url":null,"abstract":"Fuzzy rough set theory is an important approach for analyzing data uncertainty. However, the model lacks adaptive learning capabilities and cannot fit labeled data effectively in classification tasks. This study aims to introduce an adaptive learning mechanism into fuzzy rough set theory to enhance its data-fitting capability. To this end, this study seamlessly integrates fuzzy rough set theory with neural networks and proposes a novel fuzzy rough neural network model. This model adaptively learns fuzzy similarity relations in rough set models using the backpropagation algorithm. The proposed network model comprises five layers: the input, membership, fuzzy lower approximation, fully connected, and output layers. The fuzzy similarity relations between the input samples and training samples are computed in the membership layer. These relations are utilized in the fuzzy lower approximation layer to describe the degree to which the samples belong to different classes. The fuzzy rough lower approximations of the input samples are finally fused in the fully connected layer using feature weight coefficients. In the backpropagation stage, the gradient of the objective function is used to correct the fuzzy similarity relations and feature weight coefficients. This study theoretically proved that the proposed fuzzy rough network has a generalized function approximation property and can approximate any decision function. Experimental analysis showed that the proposed method is effective and performs better than most of the existing state-of-the-art algorithms.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 10","pages":"7579-7590"},"PeriodicalIF":8.7,"publicationDate":"2025-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145090077","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-26DOI: 10.1109/TSMC.2025.3598080
Qing Guo;Haoran Zhan;Zhao Wang;Tieshan Li
There exist general model uncertainty and shutoff deadzone in hydraulic-driving plant due to unknown model parameters of mechanical structure and physical feature of electrohydraulic actuator, which will degrade the motion performance and stability. In this study, a fixed-time sliding mode adaptive control is presented in 2-DOF hydraulic manipulator to address these issues. First, the dynamic manipulator with two degree-of-freedom joint driving by hydraulic servo valve is setup via Lagrangian method. Then, a sliding mode disturbance observer is designed to approximate uncertain nonlinearity to ensure the lumped uncertainties convergence in a practical finite time. Furthermore, a parametric adaptive estimation is used to estimate unknown shutoff deadzone parameters. According to backstepping technique, a fixed-time convergence control is adopted to ensure all the system state errors converge into a zero neighborhood in a constant time not related to any initial condition. Finally, the fixed-time sliding mode adaptive controller has been verified in an experimental bench of 2-DOF manipulator.
{"title":"Fixed-Time Sliding Mode Adaptive Control of Hydraulic Manipulator With Shutoff Deadzone and Uncertain Nonlinearity","authors":"Qing Guo;Haoran Zhan;Zhao Wang;Tieshan Li","doi":"10.1109/TSMC.2025.3598080","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3598080","url":null,"abstract":"There exist general model uncertainty and shutoff deadzone in hydraulic-driving plant due to unknown model parameters of mechanical structure and physical feature of electrohydraulic actuator, which will degrade the motion performance and stability. In this study, a fixed-time sliding mode adaptive control is presented in 2-DOF hydraulic manipulator to address these issues. First, the dynamic manipulator with two degree-of-freedom joint driving by hydraulic servo valve is setup via Lagrangian method. Then, a sliding mode disturbance observer is designed to approximate uncertain nonlinearity to ensure the lumped uncertainties convergence in a practical finite time. Furthermore, a parametric adaptive estimation is used to estimate unknown shutoff deadzone parameters. According to backstepping technique, a fixed-time convergence control is adopted to ensure all the system state errors converge into a zero neighborhood in a constant time not related to any initial condition. Finally, the fixed-time sliding mode adaptive controller has been verified in an experimental bench of 2-DOF manipulator.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 10","pages":"7566-7578"},"PeriodicalIF":8.7,"publicationDate":"2025-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145090073","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-22DOI: 10.1109/TSMC.2025.3595714
Mingyang Chen;Yeming Gong
While previous studies have examined how blockchain adoption affects stakeholders’ decision-making, the vertical demand information-sharing among supply chain members within a blockchain framework remains underexplored, which raises the question of how a blockchain-enabled retailer shares the demand information and how the blockchain technology may work in a supply chain. This study investigates the impacts of introducing blockchain technology on the information-sharing strategy and equilibrium choice. We find that, in the scenario of blockchain technology, information sharing may not occur when the product quality is higher since sharing may benefit the upstream and hurt the downstream. When the quality is lower, information sharing can be better off for all members given the conditions: 1) the market dispersion is larger or 2) the market dispersion is smaller and the cost is higher. In the scenario of no information-sharing, the implementation of blockchain technology is always better off for the supplier due to the increase in the wholesale price and order quantity, while conditionally benefiting the retailer if the implementation cost of blockchain technology is lower. We also find that consumer surplus is lowest in the case of no information-sharing and no blockchain because of the higher price and nontraceability of information.
{"title":"Vertical Information Sharing in a Blockchain-Enabled Supply Chain","authors":"Mingyang Chen;Yeming Gong","doi":"10.1109/TSMC.2025.3595714","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3595714","url":null,"abstract":"While previous studies have examined how blockchain adoption affects stakeholders’ decision-making, the vertical demand information-sharing among supply chain members within a blockchain framework remains underexplored, which raises the question of how a blockchain-enabled retailer shares the demand information and how the blockchain technology may work in a supply chain. This study investigates the impacts of introducing blockchain technology on the information-sharing strategy and equilibrium choice. We find that, in the scenario of blockchain technology, information sharing may not occur when the product quality is higher since sharing may benefit the upstream and hurt the downstream. When the quality is lower, information sharing can be better off for all members given the conditions: 1) the market dispersion is larger or 2) the market dispersion is smaller and the cost is higher. In the scenario of no information-sharing, the implementation of blockchain technology is always better off for the supplier due to the increase in the wholesale price and order quantity, while conditionally benefiting the retailer if the implementation cost of blockchain technology is lower. We also find that consumer surplus is lowest in the case of no information-sharing and no blockchain because of the higher price and nontraceability of information.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 10","pages":"7459-7471"},"PeriodicalIF":8.7,"publicationDate":"2025-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145089947","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-22DOI: 10.1109/TSMC.2025.3595891
Fei Zhang;Guang-Hong Yang
This article studies the pursuit–evasion game involving nonholonomic vehicles constrained by input saturation, aiming for the pursuer to intercept an evasive opponent. Unlike the previous game research neglecting the practical kinematic constraints, a coupled nonlinear system is formulated to elucidate the interaction dynamics between the players. After that, the optimal control strategies are derived by solving the Hamilton–Jacobi–Isaacs (HJI) equation linked to a special nonquadratic cost function. The Nash equilibrium analysis and finite-time capturability are conducted. To learn the optimal pursuit–evasion strategy pair, a fixed-time convergent reinforcement learning (RL) algorithm is proposed, which leverages a novel residual design to facilitate weight updates by collecting and evaluating current and historical data based on information quality. Compared with the existing RL methods that suffer from sluggish convergence due to an asymptotic learning rule and the stringent persistent excitation (PE) condition, the proposed RL relaxes the PE to an easily achievable and online verifiable finite excitation (FE) condition, allowing rapid weight convergence within a fixed period. Simulations and comparisons validate the effectiveness and superiority of the proposed method, showing a 61% reduction in convergence time in contrast to the prevailing RL schemes.
{"title":"Dynamic Historical Data-Based Reinforcement Learning for Pursuit–Evasion Games of Nonholonomic Vehicles With Input Saturation","authors":"Fei Zhang;Guang-Hong Yang","doi":"10.1109/TSMC.2025.3595891","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3595891","url":null,"abstract":"This article studies the pursuit–evasion game involving nonholonomic vehicles constrained by input saturation, aiming for the pursuer to intercept an evasive opponent. Unlike the previous game research neglecting the practical kinematic constraints, a coupled nonlinear system is formulated to elucidate the interaction dynamics between the players. After that, the optimal control strategies are derived by solving the Hamilton–Jacobi–Isaacs (HJI) equation linked to a special nonquadratic cost function. The Nash equilibrium analysis and finite-time capturability are conducted. To learn the optimal pursuit–evasion strategy pair, a fixed-time convergent reinforcement learning (RL) algorithm is proposed, which leverages a novel residual design to facilitate weight updates by collecting and evaluating current and historical data based on information quality. Compared with the existing RL methods that suffer from sluggish convergence due to an asymptotic learning rule and the stringent persistent excitation (PE) condition, the proposed RL relaxes the PE to an easily achievable and online verifiable finite excitation (FE) condition, allowing rapid weight convergence within a fixed period. Simulations and comparisons validate the effectiveness and superiority of the proposed method, showing a 61% reduction in convergence time in contrast to the prevailing RL schemes.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 10","pages":"7539-7550"},"PeriodicalIF":8.7,"publicationDate":"2025-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145090078","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：