检索结果-内蒙古大学图书馆

VFL+: Low-Coupling Vertical Federated Learning With Privileged Information Paradigm

IEEE Transactions on Emerging Topics in Computational Intelligence 2025年

作者： Dai, Wei Cui, Teng Zhang, Tong Chen, Badong China University of Mining Technology School of Information and Control Engineering Xuzhou221116 China South China University of Technology School of Computer Science and Engineering Guangzhou510006 China Xi'an Jiaotong University Institute of Artificial Intelligence and Robotics Xi'an710049 China

Vertical Federated Learning (VFL) enables the construction of models by combining clients with different features without compromising privacy. Existing VFL methods exhibit tightly coupled participant parameters, resulting in substantial interdependencies among clients during the prediction phase, which significantly hampers the model's usability. To tackle these challenges, this paper studies a VFL approach with low coupling of parameters between clients. Drawing inspiration from federated cooperation and teacher-supervised learning, we propose a low-coupling vertical federated learning with privileged information paradigm (VFL+), allowing participants to make autonomous predictions. Specifically, VFL+ treats information from other clients as privileged data during the training phase rather than the testing phase, thereby achieving independence in individual model predictions. Subsequently, this paper further investigates three typical scenarios of vertical cooperation and designs corresponding cooperative frameworks. Systematic experiments on real data sets demonstrate the effectiveness of the proposed method. © 2017 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Two-Stage Incentive Mechanism Based on a Cooperative Mode: A Stackelberg Game Approach

引用

IEEE Internet of Things Journal 2025年

作者： Wang, Long Wang, Yingjie Teng, Haojun Jiao, Xiuzhen Sun, Meimei Yang, Jishen Cai, Zhipeng Yantai University School of Computer and Control Engineering Yantai264005 China China Judicial Big Data Research Institute Co. Ltd China Naval Aviation University China Georgia State University Department of Computer Science AtlantaGA30303 United States

With the explosive growth of mobile data, Mobile Crowd Sensing (MCS) has become a popular paradigm for large-scale data collection. The difficulty of data collection and the gaps in workers’ sensing capabilities are key factors to consider in worker recruitment and task assignment. To address these issues, we designed a two-stage cooperative incentive mechanism. In the first stage, a shelving level is introduced to assess task difficulty. Tasks are divided into high-quality and low-quality groups based on quality scores, while workers are categorized into high-ability and low-ability groups based on their historical performance. The Improved Chaotic Particle Swarm Optimization (ICPSO) algorithm is then applied to generate optimal task combinations. In the second stage, we address benefit distribution among cooperating workers by introducing time-dependent rewards and employing Stackelberg game theory to analyze the optimal completion time for both high-ability and low-ability workers. This analysis determines the optimal reward distribution and ability value allocation, with proof of the existence and uniqueness of a Nash equilibrium. Comparative experiments conducted on real-world datasets demonstrate that our cooperative task mechanism outperforms existing task bundling methods, validating its rationality and effectiveness. © 2014 IEEE.

关键词： Nash equilibrium

来源：评论

学校读者我要写书评

暂无评论

Energy-Aware Task Allocation for Teams of Multi-mode Robots

arXiv

引用

arXiv 2025年

作者： Ito, Takumi Funada, Riku Sampei, Mitsuji Notomista, Gennaro Department of Systems and Control Engineering Institute of Science Tokyo Tokyo152-8550 Japan Department of Electrical and Computer Engineering University of Waterloo WaterlooONN2L 3G1 Canada

This work proposes a novel multi-robot task allocation framework for robots that can switch between multiple modes, e.g., flying, driving, or walking. We first provide a method to encode the multi-mode property of robots as a graph, where the mode of each robot is represented by a node. Next, we formulate a constrained optimization problem to decide both the task to be allocated to each robot as well as the mode in which the latter should execute the task. The robot modes are optimized based on the state of the robot and the environment, as well as the energy required to execute the allocated task. Moreover, the proposed framework is able to encompass kinematic and dynamic models of robots alike. Furthermore, we provide sufficient conditions for the convergence of task execution and allocation for both robot models. Copyright © 2025, The Authors. All rights reserved.

关键词： Constrained optimization

来源：评论

学校读者我要写书评

暂无评论

Cross representation subspace learning for multi-view clustering

引用

Expert Systems with Applications 2025年 286卷

作者： Jiang, Hongwei Ma, Wenming Dai, Jian Ding, Jianguo Tong, Xiangrong Wang, Yingjie Du, Xiaolin Jiang, Dalong Luo, Yixuan Zhang, Jinghui School of Computer and Control Engineering Yantai University YanTai264005 China Xinjiang Academy of Agricultural Sciences Urumqi830091 China Institute of Agricultural Economics and Information Xinjiang Academy of Agricultural Sciences Urumqi830091 China

In recent years, there has been growing interest in multi-view subspace clustering for handling multiple-view data. While current methodologies typically aim to capture both consistency and complementarity between views, many approaches still struggle to fully exploit the potential of multiple-view data. Moreover, most existing methods operate directly in the raw space, overlooking the complex higher-order correlations across different views. To overcome the above issues, we have proposed a new approach named CRSL (Cross Representation Subspace Learning). This approach utilizes latent and subspace self-expressive representations to learn consistency and complementarity information simultaneously in multiple-view data. Additionally, we introduce a tensor structure to explore higher-order correlations between multiple views as a way to enhance the clustering results. The validity of the CRSL approach has been confirmed by experiments conducted on eight multiple-view data sets, demonstrating its advantages in dealing with multiple-view data and providing an effective approach for multi-view clustering. © 2025 Elsevier Ltd

关键词： Higher order statistics

来源：评论

学校读者我要写书评

暂无评论

FPGA-based Real-time High-resolution Image Processing for rocket on-board processing

FPGA-based Real-time High-resolution Image Processing for ro...

引用

International Conference on computer Research and Development (ICCRD)

作者： Hui Zhou Xiaofeng Wang YuJia Xie Yongxiang Cao Yu Tang Yongchun Huo Huan Wen Xiongbo Zhao Chaoran Li Beijing Aerospace Automatic Control Institute National key Laboratory of Science and Technology on Aerospace Intelligent Control Beijing China Computer Science and Engineering Beihang University Beijing China National Superior College for Engineers Beihang University Beijing China

ISBN: (数字)9798331531881

ISBN: (纸本)9798331531898

The structure of the rocket-borne model is inherently complex, with processed images exhibiting high resolution and generating substantial amounts of data and calculations. Achieving robust real-time computing on an embedded platform poses significant challenges due to strictly limited resources, power consumption constraints, and size limitations. Our review of rocket-borne applications reveals considerable variability in the design resources of different devices, indicating a need for expanded design approaches. Upon evaluating existing methods, we identified two primary drawbacks. First, certain operators within the high-resolution target detection model are difficult to parallelize, resulting in significant inference delays that hinder the ability to meet task requirements. Although existing methods have been extended, there remains significant potential for performance enhancement in core scheduling for poor acceleration. This paper proposes an optimized architecture for the target detection algorithm accelerator designed for high-resolution images, along with a novel highly parallel data pre-processing and post-processing module implemented on FPGA to address these issues. Compared to the ARM implementation, this architecture demonstrates an improved performance of 24.64x. Furthermore, to ensure flexible application across various rocket launch scenarios, we introduce an optimization structure for convolution, pooling, and fusion operators and a multi-core expansion optimization method. This approach yields a 1.29x improvement in computing unit utilization compared to state-of-the-art multi-core scaling efforts. Finally, we assessed the accelerator architecture across multiple FPGA platforms, achieving a peak processing element utilization rate of 99.71% for a single core and layer. The overall computing efficiency, excluding the first layer, exceeded 90%. The peak computing power for the four cores reached 1638.4 GOPS, and the end-to-end computation time for

关键词： YOLO Rockets Power demand Convolution Multicore processing Reviews Scalability Accelerator architectures Real-time systems Field programmable gate arrays

来源：评论

学校读者我要写书评

暂无评论

Deep Reinforcement Learning for Job Scheduling and Resource Management in Cloud Computing: An Algorithm-Level Review

arXiv

引用

arXiv 2025年

作者： Gu, Yan Liu, Zhaoze Dai, Shuhong Liu, Cong Wang, Ying Wang, Shen Theodoropoulos, Georgios Cheng, Long The School of Control and Computer Engineering North China Electric Power University Beijing China The School of Computer Science and Technology Shandong University of Technology China The Institute of Computing Technology Chinese Academy of Sciences Beijing China The School of Computer Science University College Dublin Ireland Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen China

Cloud computing has revolutionized the provisioning of computing resources, offering scalable, flexible, and on-demand services to meet the diverse requirements of modern applications. At the heart of efficient cloud operations are job scheduling and resource management, which are critical for optimizing system performance and ensuring timely and cost-effective service delivery. However, the dynamic and heterogeneous nature of cloud environments presents significant challenges for these tasks, as workloads and resource availability can fluctuate unpredictably. Traditional approaches, including heuristic and meta-heuristic algorithms, often struggle to adapt to these real-time changes due to their reliance on static models or predefined rules. Deep Reinforcement Learning (DRL) has emerged as a promising solution to these challenges by enabling systems to learn and adapt policies based on continuous observations of the environment, facilitating intelligent and responsive decision-making. This survey provides a comprehensive review of DRL-based algorithms for job scheduling and resource management in cloud computing, analyzing their methodologies, performance metrics, and practical applications. We also highlight emerging trends and future research directions, offering valuable insights into leveraging DRL to advance both job scheduling and resource management in cloud computing. © 2025, CC BY.

关键词： Job shop scheduling

来源：评论

学校读者我要写书评

暂无评论

Dual-Mode Disturbance Observer-based Predictive control with Harmonic Current Suppression for Dual Three-Phase Permanent Magnet Synchronous Motor Drives

引用

IEEE Journal of Emerging and Selected Topics in Power Electronics 2025年

作者： Liu, Huanli Lin, Weiyang Liu, Zhitai Yu, Xinghu Buccella, Concettina Cecati, Carlo Harbin Institute of Technology Research Institute of Intelligent Control and Systems Harbin150001 China Ningbo Institute of Intelligent Equipment Technology Company Ltd Ningbo315200 China University of L'Aquila DigiPower srl Department of Information Engineering Computer Science and Mathematics L'Aquila67100 Italy

Dual three-phase permanent magnet synchronous motor drives have gained considerable interest in those applications requiring high levels of reliability. However, they often suffer from high-order harmonic currents and mismatch of motor parameters, resulting in degraded performance and reduced efficiency. Therefore, this paper proposes a dual-mode disturbance observer based model predictive current controller. Not only direct current disturbances caused by motor parameters, but also alternating current disturbances caused by high-order harmonic currents are explicitly modeled and estimated to achieve improved model accuracy. In addition, the observer parameter design is given in detail. Furthermore, a generalized predictive current controller with the integration of the proposed observer is developed to provide satisfactory harmonic current suppression capability and parameter robustness simultaneously. Proof of the stability with the proposed controller is given and several experiments and comparisons demonstrate the effectiveness of the proposed method. © 2013 IEEE.

关键词： Permanent magnets

来源：评论

学校读者我要写书评

暂无评论

On Word-of-Mouth and Private-Prior Sequential Social Learning

arXiv

引用

arXiv 2025年

作者： Da Col, Andrea Rojas, Cristian R. Krishnamurthy, Vikram The Division of Decision and Control Systems KTH Royal Institute of Technology Stockholm100 44 Sweden The School of Electrical and Computer Engineering Cornell University IthacaNY14853 United States

— Social learning provides a fundamental framework in economics and social sciences for studying interactions among rational agents who observe each other’s actions but lack direct access to individual beliefs. This paper investigates a specific social learning paradigm known as Word-of-Mouth (WoM), where a series of agents seeks to estimate the state of a dynamical system. The first agent receives noisy measurements of the state, while each subsequent agent relies solely on a degraded version of her predecessor’s estimate. A defining feature of WoM is that the final agent’s belief is publicly broadcast and adopted by all agents, in place of their own. We analyze this setting both theoretically and through numerical simulations, showing that some agents benefit from using the public belief broadcast by the last agent, while others suffer from performance deterioration. © 2025, CC BY.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Calibration-free method of recovering the time-varying absorbance profile in TDLAS by the Fourier reconstructed spectroscopy

引用

Measurement: Journal of the International Measurement Confederation 2025年 253卷

作者： Yang, Youquan Peng, Zhimin Zheng, Dao Ding, Yanjun Du, Yanjun School of Control and Computer Engineering North China Electric University Beijing102206 China State Key Laboratory of Power Systems Department of Energy and Power Engineering Tsinghua University Beijing100084 China Shanxi Research Institute of Huairou Laboratory Taiyuan030032 China

Traditional scan-direct absorption spectroscopy (scan-DAS) is susceptible to 1/f noise, limiting its ability to realize high signal-to-noise ratio (SNR) measurements. This study proposes a calibration-free and noise-immune method, named fast scan–Fourier reconstructed spectroscopy (FC-FRS), capable of improving the SNR of the single-shot absorbance spectrum and reconstructing the time-varying absorbance without decreasing the time resolution. Two validation experiments with CO2 absorbance varying over time were designed using a shock tube device within the millisecond time duration. In the first validation experiment, CO2 was gradually released during the combustion of hydrocarbons, resulting in the slowly increased CO2 absorbance. The average spectral SNR of FC-FRS reconstructed absorbance at 2100 K reached 73.51, which increased by 22.83 % compared to the original scan-DAS spectrum. The CO2 generation profile extracted from the recovered absorbance agreed well with the prediction of kinetic mechanisms. In the second experiment where the CO2 absorbance was stable behind the reflected shock at 1312 K, the average spectral SNR of the recovered absorbance line behind the reflected shock was 433.81, which is 44 % higher than that of the scan-DAS spectrum. The averaged fitted line strength from the recovered absorbance curve had a relative error of only 1.07 % compared to the HITAN2020 database. Finally, the proposed method was compared with the 2f/1f method of scan-WMS in a static CO measurement experiment, the FC-FRS showed the similar SNR but better stability over time. © 2025 Elsevier Ltd

关键词： Absorption spectroscopy

来源：评论

学校读者我要写书评

暂无评论

Hierarchical Hybrid Model Based on Deep Learning for Myocardial Infarction Detection 25

Hierarchical Hybrid Model Based on Deep Learning for Myocard...

引用

Proceedings of the 2025 4th International Conference on Big Data, Information and computer Network

作者： Mingxi Han Chi Ma Siyuan Li Baoming Pu School of Software Liaoning Technical University Huludao Liaoning China School of Computer Science and Engineering Northeastern University Shenyang Liaoning China School of Electrical and Intelligent Control Liaoning Institute of Science and Engineering Jinzhou Liaoning China

ISBN: (纸本)9798400712425

Myocardial Infarction (MI) is a major global health threat, where rapid and accurate diagnosis is essential for improving treatment outcomes. This study proposes MSRC-TransBLSTM, a deep learning-based hierarchical hybrid model for the automatic detection of MI. The model combines spatial and temporal features through a hierarchical modeling strategy: multi-layer convolutional blocks and improved MSRC modules extract and optimize spatial features, strengthening the representation of both local and global features. For temporal modeling, the Transformer Encoder captures global dependencies, while the BLSTM focuses on refining local dynamics features. Experiments on the PTB-XL dataset demonstrated the model's strong performance across key metrics (Acc = 98.68%, Sen = 97.33%, F1 = 97.43%). Compared to other models, it achieves notable improvements in accuracy and feature representation, confirming its effectiveness in MI detection.

关键词： Deep Learning

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：