检索结果-内蒙古大学图书馆

Co-Designing Transformer Architectures for Distributed Inference With Low Communication

IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS 2025年第4期36卷 717-730页

作者： Du, Jiangsu Wei, Yuanxin Ye, Shengyuan Jiang, Jiazhi Chen, Xu Huang, Dan Lu, Yutong Sun Yat Sen Univ Sch Comp Sci & Engn Guangzhou 510275 Peoples R China

Transformer models have shown significant success in a wide range of tasks. However, the massive resources required for its inference prevent deployment on a single device with relatively constrainted resources, thus leaving a high threshold of integrating their advancements. Observing scenarios such as smart home applications on edge devices and cloud deployment on commodity hardware, it is promising to distribute Transformer inference across multiple devices. Unfortunately, due to the tightly-coupled feature of Transformer model, existing model parallelism approaches necessitate frequent communication to resolve data dependencies, making them unacceptable for distributed inference, especially under relatively weak interconnection. In this paper, we propose DeTransformer, a communication-efficient distributed Transformer inference system. The key idea of DeTransformer involves the co-design of Transformer architecture to reduce the communication during distributed inference. In detail, DeTransformer is based on a novel block parallelism approach, which restructures the original Transformer layer with a single block to the decoupled layer with multiple sub-blocks. Thus, it can exploit model parallelism between sub-blocks. Next, DeTransformer contains an adaptive execution approach that strikes a trade-off among communication capability, computing power and memory budget over multiple devices. It incorporates a two-phase planning for execution, namely static planning and runtime planning. The static planning runs offline, containing a profiling procedure and a weight placement strategy before execution. The runtime planning dynamically determines the optimal parallel computing strategy from an expertly crafted search space based on real-time requests. Notably, this execution approach can adapt to heterogeneous devices by distributing workload based on devices' computing capabilities. We conduct experiments for both auto-regressive and auto-encoder tasks of Transforme

关键词： Parallel processing Transformers computational modeling Data models Planning Distributed databases Runtime Hardware Adaptation models Throughput Model parallelism distributed transformer inference deep learning system

来源：评论

学校读者我要写书评

暂无评论

Temporal Convolutional Network Approach to Secure Open Charge Point Protocol (OCPP) in Electric Vehicle Charging

引用

IEEE ACCESS 2025年 13卷 15272-15289页

作者： Benfarhat, Ikram Goh, Vik Tor Lim Siow, Chun Sheraz, Muhammad Chee Chuah, Teong Multimedia Univ Fac Engn Cyberjaya 63100 Malaysia

Securing data transactions across computer networks presents significant challenges, particularly concerning data privacy and cybersecurity threats. These issues are particularly critical in electric vehicle charging stations (EVCSs) due to the sensitive data involved, such as data breaches, unauthorized access, and privacy concerns. The primary challenge within EVCS architecture lies in defending against various cyberattacks. Several machine learning models, including convolutional neural networks, recurrent neural networks, and long short-term memory, have been employed to enhance EVCS security. However, these models often struggle to effectively handle the temporal dependencies, complexity, and scalability required for real-time threat detection. In this study, we propose a novel temporal convolutional network (TCN) to enhance EVCS security by rapidly detecting Open Charge Point Protocol (OCPP) attacks. Our model was trained and tested on the latest CICEVSE2024 dataset, which encompasses attacks targeting EVCS communication protocols, including OCPP and ISO15118, during charging and idle states. The proposed model achieved 100% accuracy in detecting binary labels and 5-class scenarios, and 93% accuracy in identifying 17-class attack types, surpassing existing approaches. Additionally, we conducted a comparative analysis with a hybrid model, demonstrating that the proposed TCN delivers superior performance with lower computational complexity, fewer parameters, a smaller model size, and shorter computation times. These features make the TCN model highly suitable for real-time EVCS applications. This work establishes a new benchmark for EVCS security, showcasing the potential of TCNs to deliver scalable and efficient solutions for safeguarding critical charging infrastructure against cyber threats.

关键词： Long short term memory computational modeling Security Electric vehicle charging Data models Accuracy Real-time systems Adaptation models Biological system modeling Scalability Temporal convolutional networks (TCNs) charge stations (EVCS) charging management systems (EVCMS) OCPP classification and detection vulnerabilities real-time systems

来源：评论

学校读者我要写书评

暂无评论

Superpixel-Based Bipartite Graph Clustering Enriched With Spatial Information for Hyperspectral and LiDAR Data

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2025年 63卷

作者： Cao, Zhe Lu, Yihang Xin, Haonan Wang, Rong Nie, Feiping Sebilo, Mathieu Northwestern Polytech Univ Sch Artificial Intelligence Opt & Elect iOPEN Xian 710072 Shaanxi Peoples R China Sorbonne Univ Inst Ecol & Environm Sci Paris IEES F-75005 Paris France

The surge in remote sensing (RS) data underscores the need for improved data diversity and processing. While integrating hyperspectral (HS) and light detection and ranging (LiDAR) data enhances analysis and addresses spectral variability, the high dimensionality, noise, and outliers inherent in hyperspectral images present significant challenges. In addition, the precise labeling required for HS makes supervised classification labor-intensive, professional-focused, and time-consuming, further motivating the development of advanced HS clustering algorithms to address these issues. Unsupervised clustering addresses the above issues but still struggles due to the underutilization of auxiliary spatial and structural information, high data dimensionality with redundant hyperspectral bands, and information divergence from heterogeneity among multimodal data. These challenges impede the effective extraction of consistent structures, undermining clustering stability and overall model performance. To address these challenges, we propose a superpixel-based bipartite graph clustering (SBGC) enriched with spatial information for hyperspectral and LiDAR data models. Our proposed method fully utilizes spatial information to construct meaningful bipartite graphs for the efficient processing of multimodal RS data. By adopting a projected clustering paradigm, our approach simultaneously clusters and reduces dimensionality, effectively eliminating redundant bands. In addition, it innovatively stacks multimodal data into tensors, thoroughly exploring the consistent structures in the low-rank space among different modalities. This reduces the heterogeneity-induced information divergence and significantly enhances clustering performance. Extensive experiments on real datasets confirm the method's effectiveness and advanced capabilities.

关键词： Hyperspectral imaging Tensors Laser radar Discrete Fourier transforms Data models Feature extraction Land surface computational modeling Clustering algorithms Bipartite graph Bipartite graphs dimensionality reduction (DR) remote sensing (RS) scalable method spatial information tensor-based clustering unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Online Credibility Assessment of Equipment Digital Twin for Discrete Manufacturing

引用

IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING 2025年 22卷 14763-14774页

作者： Lu, Han Zhang, Lin Jamal Deen, M. Wang, Kunyu Cheng, Hongbo Yang, Laurence T. Beihang Univ Hangzhou Int Innovat Inst Hangzhou 311115 Zhejiang Peoples R China Beihang Univ Sch Automat Sci & Elect Engn Beijing 100191 Peoples R China State Key Lab Intelligent Mfg Syst Technol Beijing 100854 Peoples R China Atlas Inc Hamilton L8P 0A1 ON Canada Zhengzhou Univ Sch Comp & Artificial Intelligence Zhengzhou 450001 Peoples R China Zhengzhou Univ Sch Software Engn Zhengzhou 450001 Peoples R China St Francis Xavier Univ Dept Comp Sci Antigonish NS B2G 2W5 Canada

The equipment digital twins (EDTs) for discrete manufacturing should be calibrated quickly to avoid irreversible physical damage to the equipment caused by biased control commands. Therefore, an online credibility assessment method for EDTs is urgently needed. However, existing assessment approaches consume too much time, and thus could not reveal dynamic faults in time. In this paper, the dynamic relationship between online evolution and actual applications of EDTs is investigated. Then, two steps are proposed to accelerate the assessment process significantly. One involves pre-constructing a performance-deviation-agent (PDA), and the other involves dynamically fitting the application-time-window (ATW) probability distribution. The methodology is applicable to discrete manufacturing processes. The dynamic credibility of EDT evolution process can be updated after every iteration of the model evolution. Sorting manufacturing equipment was used as a case study to demonstrate the effectiveness of this method. The time consumption was reduced by 90% compared with traditional assessment methods in the case. Note to Practitioners-The application of digital twins (DTs) for discrete manufacturing equipment typically aims to enhance the control effectiveness of the physical equipment. To ensure the effectiveness of control signals derived from DT-based simulations, the credibility of the equipment DTs must be known and guaranteed. This paper addresses the quick updating of the credibility of equipment DTs as they evolve according to real-time data collected from the equipment. The proposed methods are currently applicable to individual pieces of equipment with straightforward workflows. By adapting the training method for the error prediction model to scenarios involving multiple pieces of equipment and more complex workflows, the overall control efficacy within a workshop can be significantly improved.

关键词： Manufacturing Sorting Real-time systems Digital twins computational modeling Sensors Radiofrequency identification Artificial intelligence Training Sensitivity analysis Credibility assessment equipment digital twins manufacturing online assessment modeling simulation

来源：评论

学校读者我要写书评

暂无评论

Splitting State-Space Method for Converter-Integrated Power Systems EMT Simulations

引用

IEEE TRANSACTIONS ON POWER DELIVERY 2025年第1期40卷 584-595页

作者： Fu, Xiaopeng Wu, Wei Li, Peng Mahseredjian, Jean Wu, Jianzhong Wang, Chengshan Tianjin Univ Key Lab Smart Grid Minist Educ Tianjin 300072 Peoples R China Beijing Huairou Lab Beijing 102206 Peoples R China Polytech Montreal Dept Elect Engn Montreal PQ H3T 1J4 Canada Cardiff Univ Inst Energy Sch Engn Cardiff CF24 3AA Wales

As the utilization of power electronic-based components in power systems continues to grow, a comprehensive understanding of their dynamics becomes increasingly important for system design, control and protection analysis. To meet practical needs, the high-fidelity but time-consuming electromagnetic transient (EMT) simulations are often required. To improve the performance of these simulations, a highly efficient splitting state-space method with numerical error control is proposed that reduces the computation workload. The method employs a generic decoupling principle to split the state-space equations of the converter-integrated power system and introduces the exponential splitting formulas of multiple orders accuracy to solve and then compose the splitting state-space equations. The decoupling principle is designed based on separation of time-varying portions of the state matrix, which is realized by locating the smallest subcircuit topology that is switch state-dependent, through automatic switch grouping and switch adjacent state variables (SASV) identification. A family of exponential splitting schemes is employed to accelerate the demanding matrix exponential calculation. The splitting state-space method undergoes comprehensive testing across various cases, including a distribution network with DC load, an LLC resonant converter, a large-scale wind farm, and an MMC circuit. The accuracy of the proposed method is thoroughly evaluated, and its efficiency is validated.

关键词： Matrix converters Switches Mathematical models Integrated circuit modeling Accuracy computational modeling Topology Transient analysis State-space methods Power system dynamics Electromagnetic transient exponential integrator power converter splitting method

来源：评论

学校读者我要写书评

暂无评论

Enhancing Physical Layer Communication Security through Generative AI with Mixture of Experts

引用

IEEE WIRELESS COMMUNICATIONS 2025年第3期32卷 176-184页

作者： Zhao, Changyuan Du, Hongyang Niyato, Dusit Kang, Jiawen Xiong, Zehui Kim, Dong In Shen, Xuemin (Sherman) Letaief, Khaled B. Nanyang Technol Univ Singapore Singapore CNRS CREATE Singapore Singapore Univ Hong Kong Hong Kong Peoples R China Guangdong Univ Technol Guangzhou Peoples R China Singapore Univ Technol & Design Singapore Singapore Sungkyunkwan Univ Seoul South Korea Univ Waterloo Waterloo ON Canada Hong Kong Univ Sci & Technol Hong Kong Peoples R China

AI technologies have become increasingly adopted in wireless communications. As an emerging type of AI technologies, generative artificial intelligence (GAI) is gaining attention in communication security. Due to its powerful learning ability, GAI models have demonstrated superiority over conventional AI methods. However, GAI still has several limitations, including high computational complexity and limited adaptability. Mixture of experts (MoE) technology, which uses multiple expert models for prediction through a gate mechanism, proposes possible solutions. In this article, we first review GAI model applications in physical layer communication security, discuss limitations, and explore how MoE can help GAI overcome these limitations. Furthermore, we propose an MoE-enabled GAI framework for network optimization problems for communication security. To demonstrate the framework's effectiveness, we provide a case study in a cooperative- friendly jamming scenario. The experimental results show that the MoE-enabled framework effectively assists the GAI algorithm, solves its limitations, and enhances communication security.

关键词： Security Adaptation models Data models Artificial intelligence Physical layer computational modeling Jamming Noise Generative AI Eavesdropping

来源：评论

学校读者我要写书评

暂无评论

A Physics-Coupled Deep Learning Framework for Hydrodynamic Diffusion modeling in Watershed Systems: Integrating Spatiotemporal Networks and Environmental Constraints

引用

IEEE ACCESS 2025年 13卷 34985-35003页

作者： Jia, L. Yen, N. Pei, Y. Univ Aizu Grad Sch Comp Sci & Engn Aizu Wakamatsu Fukushima 9658580 Japan Univ Aizu Comp Sci Div Aizu Wakamatsu Fukushima 9658580 Japan

The modeling and simulation of complex spatiotemporal systems are crucial for understanding and solving multidimensional dynamical systems, particularly in earth and environmental sciences. Accurate comprehension and computational modeling of fluid transport, environmental processes, and substance diffusion depend heavily on solving governing equations. Despite significant advancements in artificial intelligence techniques, such as deep learning and neural operator methods, challenges persist regarding robustness, scalability, and adherence to physical laws in hydrodynamic systems. This paper introduces a multi-scale interdisciplinary hybrid learning framework that integrates physics-informed neural networks with neural operator-based deep learning techniques to model hydrodynamic transport processes. By incorporating convolutional neural networks for multi-scale feature extraction and implementing hard constraints to enforce physical boundary conditions, the proposed framework enhances the stability and accuracy of predictions in dynamic fluid systems. The approach facilitates efficient reconstruction of spatiotemporal characteristics and parameterized dynamics while ensuring physical consistency. Through case studies of two-dimensional solute diffusion equations, the framework demonstrates superior generalizability and robustness in addressing high-dimensional and nonlinear fluid systems. Comparative experiments with multiple baseline models highlight significant improvements in prediction accuracy, noise resistance, and adaptability to sparse datasets. The proposed method effectively addresses limitations of traditional data-driven approaches by integrating domain-specific physical principles, providing a reliable and high-fidelity solution for hydrodynamic modeling and environmental system monitoring. This research offers a transformative approach to feature extraction, dynamic representation, and physically consistent modeling of complex spatiotemporal systems.

关键词： Mathematical models Data models computational modeling Spatiotemporal phenomena Deep learning Boundary conditions Watersheds Biological neural networks Feature extraction Adaptation models Deep neural networks fluid dynamics modeling physical-coupled machine learning watershed systems environmental constraints

来源：评论

学校读者我要写书评

暂无评论

GSNN: A Neuromorphic Computing Model for the Flexible Path Planning in Various Constraint Environments

引用

IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING 2025年 22卷 1081-1099页

作者： Ruan, Haojie Chang, Yinghui Wu, Weikang Huang, Zenan Deng, Yabin Li, Leilei Luo, Hongyan He, Shan Guo, Donghui Xiamen Univ Sch Elect Sci & Engn Dept Microelect & Integrated Circuits Xiamen 361005 Peoples R China China Acad Network & Commun CETC Xiamen 361000 Peoples R China Xiamen Univ Sch Elect Sci & Engn Xiamen 361005 Peoples R China

Spiking Neural Networks (SNNs) represent a new generation of artificial neural networks that draw inspiration from biological systems. However, due to the intricate dynamics they exhibit and the discontinuity inherent in spike signals, SNNs often encounter performance limitations when addressing optimization problems. In this paper, we introduce the Graph-connected Spiking Neural Network model (GSNN), an extension of the SNN framework. The GSNN model holds the potential for integration with various existing path planning methods, rendering it applicable to a wide array of common path planning tasks. We specifically present two fundamental models within the GSNN framework. The first model employs GSNN to extract heuristic information from constrained pixel maps. This extracted data is then amalgamated with a novel sampling method, resulting in enhanced planning efficiency when compared to conventional techniques. The second model leverages GSNN to map a weighted graph, effectively utilizing plasticity methods to ascertain the shortest path within the graph. Moreover, this model facilitates path planning under diverse constraint environments, encompassing dynamic considerations, cost-awareness, and the collision dimensions of moving objects. Recognizing that the size of pixel maps or the number of nodes within weighted graphs might constrain GSNN's capabilities, we propose a partitioning strategy to address this limitation. Empirical results unequivocally demonstrate the superiority of both GSNN models in resolving static path planning problems. Furthermore, the second GSNN model demonstrates rational performance across various constrained scenarios.

关键词： Path planning Neurons Task analysis Optimization computational modeling Sampling methods Robots Spiking neural networks synaptic plasticity optimal path planning sampling-based path planning path-planning in constrained environments

来源：评论

学校读者我要写书评

暂无评论

Communication-Efficient MARL for Platoon Stability and Energy-Efficiency Co-Optimization in Cooperative Adaptive Cruise Control of CAVs

引用

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2025年第4期74卷 6076-6087页

作者： Hua, Min Chen, Dong Jiang, Kun Zhang, Fanggang Wang, Jinhai Wang, Bo Zhou, Quan Xu, Hongming Univ Birmingham Sch Engn Birmingham B15 2TT Warwick England Univ Virginia Environm Inst Charlottesville VA 22903 USA Univ Virginia Link Lab Charlottesville VA 22903 USA Univ Virginia Dept Comp Sci Charlottesville VA 22903 USA Southeast Univ Sch Automat Nanjing 210096 Peoples R China Wuhan Univ Technol Hubei Key Lab Adv Technol Automot Components Wuhan 430070 Peoples R China Automot New Technol Res Inst BYD Auto Shenzhen 518118 Peoples R China Tongji Univ Sch Automot Studies Shanghai 200070 Peoples R China

Cooperative adaptive cruise control (CACC) is a critical function that faces significant challenges in maintaining platoon stability and achieving energy efficiency, especially in real-world operations. The CACC of connected and autonomous vehicles (CAVs) based on the multi-agent reinforcement learning (MARL) algorithm is studied to optimize platoon stability and energy efficiency simultaneously. Then the effectiveness of communication information is the key to guaranteeing learning performance in real-world driving, and thus this paper has proposed a communication-efficient MARL by incorporating the quantified stochastic gradient descent (QSGD) and a binary differential consensus (BDC) method into a fully-decentralized MARL framework. We evaluate this BDC-MARL algorithm against several typical non-communicative and communicative MARL algorithms, including IA2C, FPrint, and DIAL, focusing on metrics such as platoon stability, fuel economy, and driving comfort. Our results demonstrate that BDC-MARL achieves superior energy savings, with improvements of up to 5.8%, an average velocity of 15.26 m/s, and an inter-vehicle spacing of 20.76 m. Additionally, we perform comprehensive analyses of communicative information-sharing efficiency and scalability across varying platoon sizes, further validating the practical effectiveness through real-world scenarios using data from the open-source OpenACC.

关键词： Safety Energy efficiency Scalability Robustness Protocols Optimization Learning systems Cruise control Predictive models computational modeling Connected and automated vehicles multi-agent deep reinforcement learning cooperative adaptive cruise control

来源：评论

学校读者我要写书评

暂无评论

AdderNet 2.0: Optimal AdderNet Accelerator Designs With Activation-Oriented Quantization and Fused Bias Removal-Based Memory Optimization

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS 2025年

作者： Zhang, Yunxiang Al Kailani, Omar Zhou, Bin Zhao, Wenfeng SUNY Binghamton Dept Elect & Comp Engn Binghamton NY 13850 USA ARS USDA Beltsville Agr Res Ctr Food Qual Lab Beltsville MD 20705 USA

Convolutional neural networks (CNNs) are computationally demanding due to expensive Multiply-ACcumulate (MAC) operations. Emerging neural network models, such as AdderNet, exploit efficient arithmetic alternatives like sum-of-absolute-difference (SAD) operations to replace the costly MAC operations, while still achieving competitive model accuracy as compared with the CNN counterparts. Nevertheless, existing AdderNet accelerators still face critical implementation challenges to achieve maximal hardware and energy efficiency at the cost of model inference accuracy loss. This paper presents AdderNet 2.0, an algorithm-hardware co-design framework featuring a novel Activation-Oriented Quantization (AOQ) strategy, a Fused Bias Removal (FBR) scheme for on-chip feature map memory bitwidth reduction, and optimal PE designs to improve the overall resource utilization towards optimal AdderNet accelerator designs. Multiple AdderNet 2.0 accelerator design variants were implemented on Xilinx KV-260 FPGA. Experimental results show that the INT6 AdderNet 2.0 accelerators achieve significant hardware resource and energy savings when compared to prior CNN and AdderNet designs.

关键词： Quantization (signal) Hardware Accuracy Convolutional neural networks Kernel computational modeling System-on-chip Neural networks Filters Field programmable gate arrays AdderNet activation-oriented quantization fused bias removal FPGA acceleration

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：