检索结果-内蒙古大学图书馆

arXiv 2019年

作者： Han, Yiwen Wang, Xiaofei Leung, Victor C.M. Niyato, Dusit Yan, Xueqiang Chen, Xu College of Intelligence and Computing Tianjin University Tianjin China College of Computer Science and Software Engineering Shenzhen University Shenzhen China Department of Electrical and Computer Engineering University of British Columbia Vancouver Canada School of Computer Science and Engineering Nanyang Technological University Singapore 2012 Lab of Huawei Technologies Shenzhen China School of Data and Computer Science Sun Yat-sen University Guangzhou China

Ubiquitous sensors and smart devices from factories and communities guarantee massive amounts of data and ever-increasing computing power is driving the core of computation and services from the cloud to the edge of the network. As an important enabler broadly changing people’s lives, from face recognition to ambitious smart factories and cities, artificial intelligence (especially deep learning) applications and services have experienced a thriving development process. However, due to efficiency and latency issues, the current cloud computing service architecture hinders the vision of "providing artificial intelligence for every person and every organization at everywhere". Thus, recently, a better solution is unleashing deep learning services from the cloud to the edge near to data sources. Therefore, edge intelligence, aiming to facilitate the deployment of deep learning services by edge computing, has received great attention. In addition, deep learning, as the main representative of artificial intelligence, can be integrated into edge computing frameworks to build intelligent edge for dynamic, adaptive edge maintenance and management. With regard to mutually benefited edge intelligence and intelligent edge, this paper introduces and discusses: 1) the application scenarios of both;2) the practical implementation methods and enabling technologies, namely deep learning training and inference in the customized edge computing framework;3) existing challenges and future trends of more pervasive and fine-grained intelligence. We believe that this survey can help readers to garner information scattered across the communication, networking, and deep learning, understand the connections between enabling technologies, and promotes further discussions on the fusion of edge intelligence and intelligent edge. Copyright © 2019, The Authors. All rights reserved.

关键词： Edge computing

来源：评论

学校读者我要写书评

暂无评论

CScrypt: A Compressive-Sensing-Based Encryption Engine for the Internet of Things: Demo Abstract 16

CScrypt: A Compressive-Sensing-Based Encryption Engine for t...

引用

Proceedings of the 14th ACM Conference on Embedded Network Sensor Systems CD-ROM

作者： Wanli Xue Chengwen Luo Rajib Rana Wen Hu Aruna Seneviratne School of Computer Science and Engineering University of New South Wales Australia and Data61 College of Computer Science and Software Engineering Shenzhen University China Institute of Resilient Regions University of Southern Queensland Australia Data61

ISBN: (纸本)9781450342636

Internet of Things (IoT) have been connecting the physical world seamlessly and provides tremendous opportunities to a wide range of applications. However, potential risks exist when IoT system collects local sensor data and uploads to the Cloud. The private data leakage can be severe with curious database administrator or malicious hackers who compromise the Cloud. In this demo, we solve this problem of guaranteeing the user data privacy and security using compressive sensing based cryptographic method. We present CScrypt, a compressive-sensing-based encryption engine for the Cloud-enabled IoT systems to secure the interaction between the IoT devices and the Cloud. Our system exploits the fact that each individual's biometric data can be trained to a unique dictionary which can be used as an encryption key meanwhile to compress the original data. We will demonstrate a functioning prototype of our system using live data stream when attending the conference.

关键词：

来源：评论

学校读者我要写书评

暂无评论

HFGNN: Efficient Graph Neural Networks Using Hub-Fringe Structures

HFGNN: Efficient Graph Neural Networks Using Hub-Fringe Stru...

引用

IEEE International Conference on Data Mining (ICDM)

作者： Pak Lon Ip Shenghui Zhang Xuekai Wei Tsz Nam Chan Leong Hou U SKL of Internet of Things for Smart City University of Macau Macau SAR China College of Computer Science and Software Engineering Shenzhen University Shenzhen China

ISBN: (数字)9798331506681

ISBN: (纸本)9798331506698

Existing message passing-based and transformer-based graph neural networks (GNNs) cannot satisfy requirements for learning representative graph embeddings due to restricted receptive fields, redundant message passing, and reliance on fixed aggregations. These methods face scalability and expressivity limitations from intractable exponential growth or quadratic complexity, restricting interaction ranges and information coverage across large graphs. Motivated by the analysis of long-range graph structures, we introduce a novel Graph Neural Network called Hub-Fringe Graph Neural Network (HFGNN). Our Hub-Fringe structure, drawing inspiration from the graph indexing technique known as Hub Labeling, offers a straightforward and effective approach for learning scalable graph representations while ensuring comprehensive coverage of information. HFGNN leverages this structure to enable selective propagation of relevant embeddings through a carefully designed message function. Theoretical analysis is presented to show the expressivity and scalability of the proposed method. Empirically, HFGNN exceeds standard GNNs on tasks including classification and regression, especially for large, long-range graphs where scalability and coverage matter. Ablation studies further confirm the benefits of our hub-fringe based graph neural network, including improved expressivity and scalability. The source codes is available at https://***/nick12340/HFGNN.

关键词： Training Analytical models Scalability Message passing Source coding Transformers Graph neural networks Labeling Standards Indexing

来源：评论

学校读者我要写书评

暂无评论

An End-to-End Training Method to Ensemble Bi-Linear Models for Knowledge Graph Completion

An End-to-End Training Method to Ensemble Bi-Linear Models f...

引用

International Conference on Machine Learning and Cybernetics (ICMLC)

作者： Si Cen Han Liu Chao Liu Xiaoying Zou Guoquan Dai Xizhao Wang College of Computer Science and Software Engineering Shenzhen University Shenzhen China Guangdong Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen China

The study of ensemble learning in knowledge graph embedding (KGE) shows that combining multiple individual KGE models can perform better on knowledge graph completion. However, existing KGE ensemble methods ignore the creation of model diversity because these methods independently train individual models, which are short of training interaction. To create rich model diversity, we propose a novel training method for ensemble bilinear models (EBM) for the problem of knowledge graph completion. EBM uses a weighted loss to allow individual KGE models to interact during training. In this way, the relations in the knowledge graph can be automatically modeled by the most appropriate model from the KGE individual ones. The experiments on knowledge graph completion show that EBM has richer diversity and performs better than the single bilinear model and the ensemble methods without interaction.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An Exploration of Robotic Process Automation in all Spans of Corporate Considerations 7

An Exploration of Robotic Process Automation in all Spans of...

引用

7th International Conference on Advanced Computing and Communication Systems, ICACCS 2021

作者： Balakrishnan, S. Hameed, M.Syed Shahul Venkatesan, K. Aswin, G. Sri Krishna College of Engg. and Technology Dept. of Computer Science and Business Systems Tamilnadu Coimbatore India University of Technology and Applied Sciences - sur PO BOX: 484 PC 411 SUR Oman St. Josephs College of Engineering Department of Ece Full Stack Software Developer at Zealbots in Chadura Tech Tamilnadu Tiruchirapalli India

ISBN: (纸本)9781665405201

In advance, the change of digital transformation persistently operates the function of Robotic Process Automation (RPA) and draws out all forms of corporate considerations. The RPA raises a certain kind of business measure mechanization innovation that assists thought the software programming framework which enhances the system robots or man-made artificial intelligence (AI). While RPA is a famous subject in the corporate world and this holds scholarly exploration on research work which does not have a hypothetical and theoretical investigation over RPA. They portray the trait of RPA in a given direction which furnishes the concentration for additional exploration on advancement research which includes a software solution based on routine works, organized structured information and predetermined results. This helps in investigating the reports of emerging advantages in obtaining the efficiency of RPA with its cost and speed that decreases all the possible errors. © 2021 IEEE.

关键词： Robot programming

来源：评论

学校读者我要写书评

暂无评论

Reputation and Incentive Mechanism for SDN Applications

Reputation and Incentive Mechanism for SDN Applications

引用

International Conference on Mobile Ad-hoc and Sensor Networks, MSN

作者： Yufu Wang Yuan Liu Jinqiao Hu Mingwei Zhang Xingwei Wang School of Computer Science and Engineering Northeastern University Shenyang China Software College Northeastern University Shenyang China NorthEast Yucai School Shenyang China

ISBN: (纸本)9781728105499;9781728105482

software Defined Networking (SDN) decouples the control plane from the data plane, which increases network scalability and flexibility. But malicious applications on SDN controller can cause the entire network to crash. So, we design a reputation and incentive mechanism on SDN to reduce application's malicious access. In the proposed module, first of all, the application behavior is analyzed and the malicious accesses are identified, which are used to build the reputation and incentive mechanism. Second, the analysis results of the application behavior are combined through beta probability density to obtain the reputation rating. The reward or punishment will be given based on the behavior and reputation of the application under the selected social strategy. Simulation results show that the system can accurately identify malicious behavior and reduce malicious requests, with an acceptable runtime overhead about 300 microseconds.

关键词： Control systems Static analysis Authorization Classification algorithms computer science software

来源：评论

学校读者我要写书评

暂无评论

A Novel Low Bit Rate Audio Bandwidth Extension Method

A Novel Low Bit Rate Audio Bandwidth Extension Method

引用

International Symposium on Knowledge Acquisition and Modeling, KAM

作者： Bo Hang Ruimin Hu Ye Ma National Engineering Research Center of Multimedia Software Wuhan University of China Wuhan China Mathematics and Computer Science College Xiangtan University Xiangfan China

In present communication system, high quality audio signal is supposed to be provided with low bit rate and low computational complexity. This paper proposed a novel audio coding bandwidth extension method, which can improve decoded audio quality with increasing only a few coding bits per frame and a little computational complexity. This method calculate high-frequency synthesis filter by using codebook mapping method, and transmit only quantified gain corrections in high-frequency part of multiplexing coding bit stream. The preliminary test show that this method can provide comparable audio quality with lower bit consumption and computational complexity compared to the high frequency regeneration of AMR-WB+.

关键词： Bit rate Bandwidth Computational complexity Audio coding Decoding Signal synthesis Filters Streaming media Testing Frequency

来源：评论

学校读者我要写书评

暂无评论

Large Language Models Improve Alzheimer's Disease Diagnosis Using Multi-Modality Data

Large Language Models Improve Alzheimer's Disease Diagnosis ...

引用

Medical Artificial Intelligence (MedAI), IEEE International Conference on

作者： Yingjie Feng Xiaoyin Xu Yueting Zhuang Min Zhang School of Software Zhejiang University Hangzhou China College of Biomedical Engineering & Instrument Science Zhejiang University Hangzhou China School of Computer Science Zhejiang University Hangzhou China

In diagnosing challenging conditions such as Alzheimer's disease (AD), while imaging is an important reference, non-image data such as patient information, genetic data, medication information, cognitive and memory tests also play a very important role. However, limited by the ability of artificial intelligence models to mine such information, most of the existing models only use image data, and cannot make comprehensive use of non-image data. We based our approach on an existing pretrained large language model (LLM) to enhance the model's ability to utilize non-image data on diagnosing AD. Our test results on the ADNI dataset show that the approached achieved state-of-the-art (SOTA) performance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ePoW Energy-Efficient Blockchain Consensus Algorithm for Decentralize Federated Learning System in Resource-Constrained UAV Swarm 5th

ePoW Energy-Efficient Blockchain Consensus Algorithm for De...

引用

5th International Conference on Blockchain and Trustworthy Systems, BlockSys 2023

作者： Fan, Yuting Chen, Jianguo Zhang, Longxin Li, Peiqi School of Software Engineering Sun Yat-Sen University Zhuhai519000 China College of Computer Science Hunan University of Technology Zhuzhou412007 China

ISBN: (纸本)9789819981038

With the proliferation of Unmanned Aerial Vehicles (UAV) and UAV swarms, there has been growing interest in using them for collaborative computing tasks. Blockchain-based Federated learning (BFL) is an excellent approach for training artificial intelligence models in UAV swarms, providing benefits such as privacy protection, trusted computing, node autonomy, and low communication overhead. However, due to limited battery life and processing power, UAVs require energy-efficient solutions that can handle blockchain consensuses in decentralized BFL manners. To address such challenges, we propose an energy-efficient Proof of Work (ePoW) consensus algorithm for the resource-constrained UAV swarms, which to achieve fault tolerance and data integrity while minimizing energy consumption. At the commencement of the ePoW protocol, the UAV nodes perform computation and dissemination of their engagement metrics to facilitate the selection of participating nodes in the ePoW competition. Following a successful BFL global generation event, the dynamic difficulty adjustment mechanism collaborates with the engagement metrics to identify the most reliable node and mitigate resource consumption during the consensus process. Our ePow algorithm is designed to work with BFL systems in a UAV swarm, where each UAV device acts as a node in the BFL system. We evaluate the algorithm’s performance in various scenarios and found that it achieves high accuracy while consuming significantly less energy compared to existing consensus algorithms. Our proposed approach presents a promising solution for energy-efficient decentralized BFL systems in UAV swarms. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

Usefulness of high-level parallel compositions in genomics 32

Usefulness of high-level parallel compositions in genomics

引用

32nd European Modeling and Simulation Symposium, EMSS 2020

作者： Rossainz-López, Mario Zúñiga-Herrera, Sarahí Pineda, Iván Olmos Pineda-Torres, Ivo Capel-Tuñón, Manuel Faculty of Computer Science Autonomous University of Puebla San Claudio Avenue and South 14th Street City San Manuel Puebla Puebla72000 Mexico Software Engineering Department College of Informatics and Telecommunications ETSIIT University of Granada Daniel Saucedo Aranda s/n Granada18071 Spain

ISBN: (纸本)9788885741454

This work shows the use of parallel objects to build High Level Parallel Compositions or HLPC and their usefulness in genomics through four case studies related to sequencing DNA chains. The first two case studies are combinatorial optimization problems: grouping fragments of DNA sequences and the parallel exhaustive search (PES) of RNA strings that help the sequence and assembly of DNAs in the construction of gnomes. The third case study shows the implementation of a Convolutional Neuronal Network as a Parallel Object Composition to solve the problem of the recognition of DNA sequences from a database with 4 types of hepatitis C virus (type 1, 2, 3 and 6). The results of this classification were obtained in terms of percentages of training precision and validation precision. The fourth and final case study shows the problem of sequence typing (STP) as a form of DNA sequence classification. It is particularized in a proposal for a parallel solution to find conserved regions of sequences that help discriminate between different types of hepatitis C virus, through the creation of a decision tree using HLPC. We show the algorithms that solves these problems using modeling and parallel simulation, their design and implementation as HLPC and the performance metrics in their parallel execution using multicores, video accelerator card and CPU-SET or processors with shared-distributed memory. © 2020 The Authors.

关键词： DNA sequences

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：