检索结果-南通市图书馆

EG-STC: An Efficient Secure Two-Party Computation Scheme Based on Embedded GPU for Artificial Intelligence Systems

Computers, Materials & Continua 2024年第6期79卷 4021-4044页

作者： zhenjiang dong Xin Ge Yuehua Huang Jiankuo dong Jiang Xu School of Computer Science Nanjing University of Posts and TelecommunicationsNanjing210023China School of Computer Nanjing University of Information Science and TechnologyNanjing210044China

This paper presents a comprehensive exploration into the integration of Internet of Things(IoT),big data analysis,cloud computing,and Artificial Intelligence(AI),which has led to an unprecedented era of *** delve into the emerging trend of machine learning on embedded devices,enabling tasks in resource-limited ***,the widespread adoption of machine learning raises significant privacy concerns,necessitating the development of privacy-preserving *** such technique,secure multi-party computation(MPC),allows collaborative computations without exposing private *** its potential,complex protocols and communication interactions hinder performance,especially on resource-constrained *** to enhance efficiency have been made,but scalability remains a *** the success of GPUs in deep learning,lever-aging embedded GPUs,such as those offered by NVIDIA,emerges as a promising ***,we propose an Embedded GPU-based Secure Two-party Computation(EG-STC)framework for Artificial Intelligence(AI)*** the best of our knowledge,this work represents the first endeavor to fully implement machine learning model training based on secure two-party computing on the Embedded GPU *** experimental results demonstrate the effectiveness of *** an embedded GPU with a power draw of 5 W,our implementation achieved a secure two-party matrix multiplication throughput of 5881.5 kilo-operations per millisecond(kops/ms),with an energy efficiency ratio of 1176.3 kops/ms/***,leveraging our EG-STC framework,we achieved an overall time acceleration ratio of 5–6 times compared to solutions running on server-grade *** solution also exhibited a reduced runtime,requiring only 60%to 70%of the runtime of previously best-known methods on the same *** summary,our research contributes to the advancement of secure and efficient machine learning implementations on resource-constrained embedded

关键词： Secure two-party computation embedded GPU acceleration privacy-preserving machine learning edge computing

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

大模型知识管理系统

引用

中兴通讯技术 2024年第2期30卷 63-71页

作者：周扬蔡霈涵董振江南京邮电大学南京210023

企业知识管理至关重要,而传统企业知识管理系统存在构建成本高、知识利用率低的问题。提出了基于大模型检索增强生成(RAG)技术构建企业知识管理系统的方案。首先介绍了整体方案架构、业务流程与4类知识构建技术,然后重点介绍了检索前处... 详细信息

企业知识管理至关重要,而传统企业知识管理系统存在构建成本高、知识利用率低的问题。提出了基于大模型检索增强生成(RAG)技术构建企业知识管理系统的方案。首先介绍了整体方案架构、业务流程与4类知识构建技术,然后重点介绍了检索前处理、知识检索、检索后处理等全流程知识检索技术,并设计了全面的测评框架。经过实践检验,该方案具有知识构建效率高且成本低、意图理解精确、知识检索准确等特点与优势。

关键词： RAG 知识管理系统大模型知识工程

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Super-plasticity of Zr_(64.80)Cu_(14.85)Ni_(10.35)Al_(10) bulk metallic glass at room temperature

引用

Chinese Science Bulletin 2008年第3期53卷 465-468页

作者： TAO PingJun YANG YuanZheng BAI XiaoJun XIE ZhiWei CHEN XianCao dong zhenjiang Wen JianGuo Faculty of Materials and Energy Guangdong University of Technology Guangzhou 510006 China College of Materials Science and Engineering Shenzhen University Shenzhen 518060 China

Generally,bulk metallic glasses(BMGs)exhibit a very limited plastic deformation under a compression load at room temperature,often less than 2% before *** this letter,through an appropriate choice of BMGs' composition,an amorphous rod of Zr64.80Cu14.85Ni10.35Al10 with a diameter of 2 mm was prepared by using copper mold suction casting.X-ray diffraction and differential scanning calorimetry were utilized to determine its structure and thermal stability,and the uniaxial compression test was adopted to study its plastic deformation behavior at room temperature *** results showed that the glass transition temperature and onset temperature of the exothermic reaction of the amorphous rod were 646 and 750 K,respectively,and its micro-hardness was 594.7 *** com-pression,when the engineering strain and engineering stress arrived at 9.05% and 1732 MPa,respec-tively,i.e.,the true strain and true stress reached 9.42% and 1560 MPa,respectively,the amorphous rod started to *** yielding,with the increase of load,the strain increased and the glass rod ulti-mately were compressed into flake-like *** the maximum engineering strain was larger than 70%,i.e.,the maximum true strain exceeded by 120%,the amorphous specimen was not fractured,indicating that it has super-plasticity at room *** the appropriate choice of compo-sition and optimization of the technological process,flexible BMG with super-plasticity at room tem-perature could be produced.

关键词：大体积金属玻璃单轴向压力试验超塑性锌铜镍铝合金

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

大模型数据扩增技术在输电线路智能监控中的应用研究

引用

邮电设计技术 2024年第9期 31-37页

作者：杨勇万超伦马建友赵文杰董振江江苏讯汇科技股份有限公司江苏南京211316 南京邮电大学江苏南京210003

针对复杂场景下输电线路可视化监控中目标识别准确率不高、误检误报率较高的问题,提出了一种结合大模型的数据扩增和YOLOv10优化算法来提升复杂场景下目标识别准确率的实现方法。首先,针对样本数据数量短缺的问题,基于Stable-diffusion... 详细信息

针对复杂场景下输电线路可视化监控中目标识别准确率不高、误检误报率较高的问题,提出了一种结合大模型的数据扩增和YOLOv10优化算法来提升复杂场景下目标识别准确率的实现方法。首先,针对样本数据数量短缺的问题,基于Stable-diffusion大模型进行数据扩增,丰富并增加了样本数量。其次,针对训练样本数量有限的情况,对YOLOv10算法进行改进,进一步强化图片特征的提取算法,优化目标识别算法,提升复杂自然场景下的目标识别准确率及性能。最终的实验结果表明,与现有的实现方法相比,针对复杂场景的输电线路的可视化监控,对目标识别的准确率从原有的52.6%提升至54.3%。

关键词：数据扩增目标识别 YOLO

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Improving Performance of Cloud Computing and Big Data Technologies and Applications

引用

ZTE Communications 2014年第4期12卷 1-2页

作者： zhenjiang dong the Cloud Computing and IT Research Institute of ZTE Corporation Chinese Association for Artificial Intelligence Nanjing University of Science and Technology China Computer Federation

Cloud computing technology is changing the development and usage patterns of IT infrastructure and applications. Virtualized and distributed systems as well as unified management and scheduling has greatly im proved computing and storage. Management has become easier, andOAM costs have been significantly reduced. Cloud desktop technology is develop ing rapidly. With this technology, users can flexibly and dynamically use virtual ma chine resources, companies＇ efficiency of using and allocating resources is greatly improved, and information security is ensured. In most existing virtual cloud desk top solutions, computing and storage are bound together, and data is stored as im age files. This limits the flexibility and expandability of systems and is insufficient for meetinz customers＇ requirements in different scenarios.

关键词： Improving Performance of Cloud Computing and Big Data Technologies and Applications HBase

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

AFSTGCN:Prediction for multivariate time series using an adaptive fused spatial-temporal graph convolutional network

引用

Digital Communications and Networks 2024年第2期10卷 292-303页

作者： Yuteng Xiao Kaijian Xia Hongsheng Yin Yu-dong Zhang zhenjiang Qian Zhaoyang Liu Yuehan Liang Xiaodan Li School of Information and Control Engineering China University of Mining and TechnologyXuzhou221116China Department of Informatics University of LeicesterLeicesterLE17RHUK The Affiliated Changshu Hospital of Soochow University Changshu215500China School of Computer Science and Engineering Changshu Institute of TechnologySuzhou215500China

The prediction for Multivariate Time Series(MTS)explores the interrelationships among variables at historical moments,extracts their relevant characteristics,and is widely used in finance,weather,complex industries and other ***,it is important to construct a digital twin ***,existing methods do not take full advantage of the potential properties of variables,which results in poor predicted *** this paper,we propose the Adaptive Fused Spatial-Temporal Graph Convolutional Network(AFSTGCN).First,to address the problem of the unknown spatial-temporal structure,we construct the Adaptive Fused Spatial-Temporal Graph(AFSTG)***,we fuse the spatial-temporal graph based on the interrelationship of spatial ***,we construct the adaptive adjacency matrix of the spatial-temporal graph using node embedding ***,to overcome the insufficient extraction of disordered correlation features,we construct the Adaptive Fused Spatial-Temporal Graph Convolutional(AFSTGC)*** module forces the reordering of disordered temporal,spatial and spatial-temporal dependencies into rule-like *** dynamically and synchronously acquires potential temporal,spatial and spatial-temporal correlations,thereby fully extracting rich hierarchical feature information to enhance the predicted *** on different types of MTS datasets demonstrate that the model achieves state-of-the-art single-step and multi-step performance compared with eight other deep learning models.

关键词： Adaptive adjacency matrix Digital twin Graph convolutional network Multivariate time series prediction Spatial-temporal graph

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Technical Analysis of Network Plug-in Flannel for Containers

引用

ZTE Communications 2017年第4期15卷 43-46页

作者： YANG Yong dong Xiugang dong zhenjiang Nanjing R&D Center ZTE Corporation

The development of cloud computing has made container tec-hnology a hot research issue in recent years. The container technology provides a basic support for micro service architecture, while container networking plays an important role in application of the container technology. In this paper, we study the technical implementation of the Flannel module, a network plug-in for Docker containers, including its func tions, implementation principle, utilization, and performance. The performance of Flannel in different modes is further tested and analyzed in real application scenarios.

关键词： container technology Docker network plug.in Flannel

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Key Technologies and Application of Edge Computing

引用

ZTE Communications 2017年第2期15卷 26-34页

作者： TU Yaofeng dong zhenjiang YANG Hongzhang Nanjing R&D Center ZTE Corporation Nanjing 210021 China

Cloud computing faces a series of challenges,such as insufficient bandwidth,unsatisfactory real-time,privacy protection,and energy *** overcome the challenges,edge computing *** computing refers to a process where the open platform that converges the core capabilities of networks,computing,storage,and applications provides intelligent services at the network edge near the source of the objects or data to meet the critical requirements for agile connection,real-time services,data optimization,application intelligence,security and privacy protection of industry *** computing consists of three elements:edge,computing,and *** computing and the Internet of Things(IoT)mutually create,and edge computing and cloud computing complement each *** the architecture of edge computing,resources are distributed to the edge nodes,and therefore the storage system is near users while the computation function is near *** this way,the stress on the backbone network can be *** this architecture,the existing key technologies for computation,networks,and storage will change ***’s edge computing solutions can ensure the service quality of operators and greatly enhance the experience of mobile users.

关键词： edge computing cloud computing IoT

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Consistency between domain wall oscillation modes and spin wave modes in nanostrips

引用

Chinese Physics B 2024年第6期33卷 511-516页

作者：董新伟吴振江 Department of Physics and Institute of Theoretical Physics and Astrophysics Xiamen UniversityXiamen 361005China

Investigations on domain wall(DW) and spin wave(SW) modes in a series of nanostrips with different widths and thicknesses have been carried out using micromagnetic simulation. The simulation results show that the frequencies of SW modes and the corresponding DW modes are consistent with each other if they have the same node number along the width direction. This consistency is more pronounced in wide and thin nanostrips, favoring the DW motion driven by *** analysis of the moving behavior of a DW driven by SWs is also carried out. The average DW speed can reach a larger value of ~ 140 m/s under two different SW sources. We argue that this study is very meaningful for the potential application of DW motion driven by SWs.

关键词： micromagnetic simulation domain wall spin wave

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Human Motion Recognition Based on Incremental Learning and Smartphone Sensors

引用

ZTE Communications 2016年第B6期14卷 59-66页

作者： LIU Chengxuan dong zhenjiang XIE Siyuan PEI Ling Shanghai Jiao Tong University Shanghai 200240 China ZTE Corporation Shenzhen 518057 China

Batch processing mode is widely used in the training process of human motiun recognition. After training, the motion elassitier usually remains invariable. However, if the classifier is to be expanded, all historical data must be gathered for retraining. This consumes a huge amount of storage space, and the new training process will be more complicated. In this paper, we use an incremental learning method to model the motion classifier. A weighted decision tree is proposed to help illustrate the process, and the probability sampling method is also used. The resuhs show that with continuous learning, the motion classifier is more precise. The average classification precision for the weighted decision tree was 88.43% in a typical test. Incremental learning consumes much less time than the batch processing mode when the input training data comes continuously.

关键词： human motion recognition ineremental learning mappingfunction weighted decision tree probability sampling

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

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：