检索结果-南通市图书馆

Towards robust neural networks via a global and monotonically decreasing robustness training strategy

Frontiers of Information Technology & Electronic Engineering 2023年第10期24卷 1375-1389页

作者： Zhen LIANG Taoran WU Wanwei LIU Bai XUE Wenjing YANG Ji WANG Zhengbin PANG Institute for Quantum Information&State Key Laboratory of High Performance Computing National University of Defense TechnologyChangsha 410073China State Key Laboratory of computer Science Institute of software Chinese Academy of SciencesBeijing 100190China School of computer Science and Technology University of Chinese Academy of SciencesBeijing 100190China College of computer Science and Technology National University of Defense TechnologyChangsha 410073China Key Laboratory of software Engineering for Complex systems National University of Defense TechnologyChangsha 410073China

Robustness of deep neural networks(DNNs)has caused great concerns in the academic and industrial communities,especially in safety-critical *** of verifying whether the robustness property holds or not in certain neural networks,this paper focuses on training robust neural networks with respect to given ***-of-the-art training methods,interval bound propagation(IBP)and CROWN-IBP,perform well with respect to small perturbations,but their performance declines significantly in large perturbation cases,which is termed“drawdown risk”in this ***,drawdown risk refers to the phenomenon that IBPfamily training methods cannot provide expected robust neural networks in larger perturbation cases,as in smaller perturbation *** alleviate the unexpected drawdown risk,we propose a global and monotonically decreasing robustness training strategy that takes multiple perturbations into account during each training epoch(global robustness training),and the corresponding robustness losses are combined with monotonically decreasing weights(monotonically decreasing robustness training).With experimental demonstrations,our presented strategy maintains performance on small perturbations and the drawdown risk on large perturbations is alleviated to a great *** is also noteworthy that our training method achieves higher model accuracy than the original training methods,which means that our presented training strategy gives more balanced consideration to robustness and accuracy.

关键词： Robust neural networks Training method Drawdown risk Global robustness training Monotonically decreasing robustness

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

Approximation Designs for Energy Harvesting Relay Deployment in Wireless Sensor Networks

引用

Journal of computer Science & Technology 2022年第4期37卷 779-796页

作者：王怿刘忆雪朱舜佳高晓沨田臣 School of Modern Posts Nanjing University of Posts and TelecommunicationsNanjing 210003China State Key Laboratory for Novel software Technology Nanjing UniversityNanjing 210023China ByteDance inc. Shanghai 201100China Key Laboratory of Artificial Intelligence of Ministry of Education Department of Computer Science and EngineeringShanghai Jiao Tong UniversityShanghai 200240China

Energy harvesting technologies allow wireless devices to be recharged by the surrounding environment, providing wireless sensor networks (WSNs) with higher performance and longer lifetime. However, directly building a wireless sensor network with energy harvesting nodes is very costly. A compromise is upgrading existing networks with energy harvesting technologies. In this paper, we focus on prolonging the lifetime of WSNs with the help of energy harvesting relays (EHRs). EHRs are responsible for forwarding data for sensor nodes, allowing them to become terminals and thus extending their lifetime. We aim to deploy a minimum number of relays covering the whole network. As EHRs have several special properties such as the energy harvesting and depletion rate, it brings great research challenges to seek an optimal deployment strategy. To this end, we propose an approximation algorithm named Effective Relay Deployment Algorithm, which can be divided into two phases: disk covering and connector insertion using the partitioning technique and the Steinerization technique, respectively. Based on probabilistic analysis, we further optimize the performance ratio of our algorithm to (5 + 6/K) where K is an integer denoting the side length of a cell after partitioning. Our extensive simulation results show that our algorithm can reduce the number of EHRs to be deployed by up to 45% compared with previous work and thus validate the efficiency and effectiveness of our solution.

关键词： approximation algorithm constraint relay deployment energy harvesting wireless sensor network

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

What’s Missing in Agile Hardware Design? Verification!

引用

Journal of computer Science & Technology 2023年第4期38卷 735-736页

作者： Babak Falsafi Parallel systems Architecture Laboratory Institute of Computer and Communication SciencesSchool of Computer andCommunication SciencesEcole Polytechnique Fédérale de LausanneLausanneCH-1015Switzerland

Agile hardware design is an approach to developing hardware systems that draws inspiration from the princ.ples and practices of agile software *** emphasizes collaboration,flexibility,iterative development,and quick adaptation to changing *** agile hardware design,the focus is on delivering functionalhardware systems in shorter development cycles while maintaining high-quality and customer *** particular,agile hardware design is of great interest in the open-source hardware ***-sourcehardware development—such as RISC-V—is at the forefront of initiatives to democratize hardware and drive innovation in chip design *** design is instrumental for the RISC-V community because it supportsrapid iteration,accommodates the evolving RISC-V standard and the addition of custom extensions,improvescommunity collaboration and time-to-market,and addresses the design challenges associated with complex architectural features.

关键词： hardware agile architectural

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

Design and Application of PCE-Oriented Multi-agent software Framework and Agent Communication Module

引用

Journal of Donghua University(English Edition) 2010年第2期27卷 258-262页

作者：刘发贵张功胜林俊郑兆妙 School of computer Science & Engineering South China University of Technology China great wall computer Shenzhen Co. LtdRm501Greatwall BuildingKefa RoadScience&Industry Park Nanshan District

With the improvement of mobile equipment performance and development of Pervasive Computing,interactive computational applications such as Multi-Agent (MA) systems in Pervasive Computing Environments (PCE) become more and more prevalent. Many applications in PCE require Agent communication,manual control,and diversity of devices. Hence system in PCE must be designed flexible,and optimize the use of network,storage and computing resources. However,traditional MA software framework cannot completely adapt to these new features. A new MA software framework and its Agent Communication Modules to solve the problem brought by PCE was proposed. To describe more precisely,it presents Wright/ADL (Architecture Description Language) description of the new framework. Then,it displays an application called AI Eleven based on this new framework. AI Eleven achieves Agent-Agent communication and good collaboration for a task. Two experiments on AI Eleven will demonstrate the new framework's practicability and superiority.

关键词：弥漫的计算多代理人(麻省) 框架建筑学描述语言(ADL )

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

Deciphering urban traffic impacts on air quality by deep learning and emission inventory

引用

Journal of Environmental Sciences 2023年第2期124卷 745-757页

作者： Wenjie Du Lianliang Chen Haoran Wang Ziyang Shan Zhengyang Zhou Wenwei Li Yang Wang School of software Engineering University of Science and Technology of ChinaHefei 230026China Suzhou Institute for Advanced Research University of Science and Technology of ChinaSuzhou 215123China Alibaba inc. Hangzhou 310052China School of computer Science and Technology University of Science and Technology of ChinaHefei 230026China CAS Key Laboratory of Urban Pollutant Conversion Department of environmental science and EngineeringUniversity of Science and Technology of ChinaHefei 230026China USTC-CityU Joint Advanced Research Center Suzhou 215123China

Air pollution is a major obstacle to future sustainability,and traffic pollution has become a large drag on the sustainable developments of future ***,combined with the large volume of real-time monitoring data,we propose a deep learning model,iDeepAir,to predict surface-level PM2.5 concentration in Shanghai megacity and link with MEIC emission inventory creatively to decipher urban traffic impacts on air *** model exhibits high-fidelity in reproducing pollutant concentrations and reduces the MAE from 25.355μg/m^(3) to 12.283μg/m^(3) compared with other *** identifies the ranking of major factors,local meteorological conditions have become a nonnegligible ***-wise relevance propagation(LRP)is used here to enhance the interpretability of the model and we visualize and analyze the reasons for the different correlation between traffic density and PM_(2.5) concentration in various regions of ***,As the strict and effective industrial emission reduction measurements implementing in China,the contribution of urban traffic to PM_(2.5) formation calculated by combining MEIC emission inventory and LRP is gradually inc.easing from 18.03%in 2011 to 24.37% in 2017 in Shanghai,and the impact of traffic emissions would be ever-prominent in 2030 according to our *** also infer that the promotion of vehicular electrification would achieve further alleviation of PM_(2.5) about 8.45% by 2030 *** insights are of great significance to provide the decision-making basis for accurate and high-efficient traffic management and urban pollution control,and eventually benefit people’s lives and high-quality sustainable developments of cities.

关键词： PM_(2.5)concentration forecast Traffic emissions Deep learning Attention mechanism New energy vehicles

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

DPPS: A novel dual privacy-preserving scheme for enhancing query privacy in continuous location-based services

引用

Frontiers of computer Science 2023年第5期17卷 197-205页

作者： Long LI Jianbo HUANG Liang CHANG Jian WENG Jia CHEN Jingjing LI College of Cyber Security Jinan UniversityGuangzhou 510632China Guangxi Key Laboratory of Trusted software Guilin University of Electronic TechnologyGuilin 541004China Department of computer Applications Guilin University of Technology at NanningNanning 530000China

Sinc. smartphones embedded with positioning systems and digital maps are widely used,location-based services(LBSs)are rapidly growing in popularity and providing unprecedented convenience in people’s daily lives;however,they also cause great concern about privacy *** particular,location queries can be used to infer users’sensitive private information,such as home addresses,places of work and appointment ***,many schemes providing query anonymity have been proposed,but they typically ignore the fact that an adversary can infer real locations from the correlations between consecutive locations in a continuous *** address this challenge,a novel dual privacy-preserving scheme(DPPS)is proposed that inc.udes two privacy protection ***,to prevent privacy disclosure caused by correlations between locations,a correlation model is proposed based on a hidden Markov model(HMM)to simulate users’mobility and the adversary’s prediction ***,to provide query probability anonymity of each single location,an advanced k-anonymity algorithm is proposed to construct cloaking regions,in which realistic and indistinguishable dummy locations are *** validate the effectiveness and efficiency of DPPS,theoretical analysis and experimental verification are further performed on a real-life dataset published by Microsoft,i.e.,GeoLife dataset.

关键词： location-based services privacy-preserving hidden Markov model k-anonymity query probability

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

Recent advances in connected and automated vehicles

引用

Journal of Traffic and Transportation Engineering(English Edition) 2019年第2期6卷 109-131页

作者： David Elliott Walter Keen Lei Miao Mahle Filter systems North America inc. Department of Engineering Technology Middle Tennessee State University

Connected and automated vehicle(CAV) is a transformative technology that has great potential to change our daily life. Therefore, CAV related research has been advanced significantly in recent years. This paper does a comprehensive review on five selected subjects that lie in the heart of CAV research:(i) inter-CAV communications;(ii) security of CAVs;(iii) intersection control for CAVs;(iv) collision-free navigation of CAVs; and(v)pedestrian detection and protection. It is believed that these topics are essential to ensure the success of CAVs and need to be better understood. For inter-CAV communications, this paper focuses on both Dedicated Short Range Communications(DSRC) and the future 5 G cellular technologies; for security of CAVs, this paper discusses both passive and active attacks and the existing solutions; for intersection control, this paper summarizes the pros and cons of both centralized and decentralized approaches; for collision avoidance, this paper concentrates on four subareas: maneuverability, vehicle networking, control confliction, and motorcycles; for pedestrian detection, this paper covers sensor, radar, and computer vision based approaches. Under each topic, this paper not only shows the stateof-the-art, but also unveils potential future research directions. By establishing connections among these subjects, this paper shows how they interact with each other and how they can be integrated into a seamless user experience. It is believed that the literature covered and conclusions drawn in this paper are very helpful to CAV researchers, application engineers, and policy makers.

关键词： Connected and automated vehicles Autonomous vehicles Intelligent transportation systems Driver-less cars

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

Behavior modeling and verification of movement authority scenario of Chinese Train Control System using AADL

引用

Science China(Information Sciences) 2015年第11期58卷 125-144页

作者： AHMAD Ehsan DONG YunWei LARSON Brian Lü JiDong TANG Tao ZHAN NaiJun School of computer Science Northwestern Polytechnical University State Key Laboratory of computer Science Institute of Software Chinese Academy of Sciences Computing and Information systems Kansas State University State Key Laboratory of Rail Traffic Control and Safety Beijing Jiaotong University

Train control systems like most digital controllers are, by definition, hybrid systems as they interact with or try to control some aspects of the physical world. Detailed behavior modeling with constraints specification and formal verification, required for reliability prediction, is a great challenge for hybrid system *** control systems further intensify this challenge with extensive interaction between computing units and their physical environment and their mutual dependence on each other. In this paper, we investigate behavior modeling and formal verification of Chinese Train Control System Level 3(CTCS-3) using Architectural Analysis & Design Language(AADL) to cope with this challenge. AADL is an architecture description language for embedded systems and is based on model-based engineering paradigm. Along with structural modeling of embedded systems using the core language constructs, AADL also provides support for language extension through annex sublanguages. In system requirements specification document, the behavior of the CTCS-3 is specified as a set of basic operation scenarios that cooperate with each other to achieve safe and secure functionality of *** Authority(MA) scenario, explored in this paper, is considered as a basic and most crucial scenario to prevent trains from colliding with each other. The detailed discrete behavior of control system is modeled and verified using the Behavior Language for Embedded systems with software(BLESS) annex sublanguage of AADL, and the continuous behavior of train with the cyber–physical interaction(communication between train and control system) is modeled using the Hybrid annex sublanguage. The behavior of the MA scenario at system level is verified using the Hybrid Hoare Logic theorem prover. Behavior constraints are specified as assertions using first-order logic formulas augmented with a simple temporal operator.

关键词： AADL behavior modeling BLESS annex CTCS hybrid annex train control system

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

A subtree-based approach to failure detection and protection for multicast in SDN

引用

Frontiers of Information Technology & Electronic Engineering 2016年第7期17卷 682-700页

作者： Vignesh RENGANATHAN RAJA Chung-Horng LUNG Abhishek PANDEY Guo-ming WEI Anand SRINIVASAN Department of systems and computer Engineering Carleton UniversityOttawaOntario KIS 5B6Canada School of computer Science Carleton UniversityOttawaOntario KIS 5B6Canada Eion inc. OttawaOntario K2K 2E3Canada

software-defined networking(SDN) has received tremendous attention from both industry and *** centralized control plane in SDN has a global view of the network and can be used to provide more effective solutions for complex problems,such as traffic *** study is motivated by recent advancement in SDN and inc.easing popularity of multicasting *** propose a technique to inc.ease the resiliency of multicasting in SDN based on the subtree protection *** is a group communication technology,which uses the network infrastructure efficiently by sending the data only once from one or multiple sources to a group of receivers that share a common *** applications,e.g.,live video streaming and video conferencing,become popular,but they are delay-sensitive *** in an ongoing multicast session can cause packet losses and delay,which can significantly affect quality of service(Qo S).In this study,we adapt a subtree-based technique to protect a multicast tree constructed for Open Flow switches in *** proposed algorithm can detect link or node failures from a multicast tree and then determines which part of the multicast tree requires changes in the flow table to recover from the *** a centralized controller in SDN,the backup paths can be created much more effectively in comparison to the signaling approach used in traditional multiprotocol label switching(MPLS) networks for backup paths,which makes the subtree-based protection mechanism *** also implement a prototype of the algorithm in the POX controller and measure its performance by emulating failures in different tree topologies in Mininet.

关键词： software-defined networks(SDNs) Open Flow Multicast tree Protection POX controller Mininet Multiprotocol label switching(MPLS)

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

FDNet:A Deep Learning Approach with Two Parallel Cross Encoding Pathways for Precipitation Nowcasting

引用

Journal of computer Science & Technology 2023年第5期38卷 1002-1020页

作者：闫碧莹杨超陈峰 Kohei Takeda Changjun Wang University of Chinese Academy of Sciences Beijing 100049China Institute of software Chinese Academy of SciencesBeijing 100190China School of Mathematical Sciences Peking UniversityBeijing 100871China Peng Cheng Laboratory Shenzhen 518052China Guiyang Academy of Information Technology Guiyang 550081China NTT DATA Corporation Tokyo 163-8001Japan NTT DATA Institute of Management Consulting inc. Tokyo 163-8001Japan

With the goal of predicting the future rainfall intensity in a local region over a relatively short period time,precipitation nowcasting has been a long-time scientific challenge with great social and economic *** radar echo extrapolation approaches for precipitation nowcasting take radar echo images as input,aiming to generate future radar echo images by learning from the historical *** effectively handle complex and high non-stationary evolution of radar echoes,we propose to decompose the movement into optical flow field motion and morphologic *** this idea,we introduce Flow-Deformation Network(FDNet),a neural network that models flow and deformation in two parallel cross *** flow encoder captures the optical flow field motion between consecutive images and the deformation encoder distinguishes the change of shape from the translational motion of radar *** evaluate the proposed network architecture on two real-world radar echo *** model achieves state-of-the-art prediction results compared with recent *** the best of our knowledge,this is the first network architecture with flow and deformation separation to model the evolution of radar echoes for precipitation *** believe that the general idea of this work could not only inspire much more effective approaches but also be applied to other similar spatio-temporal prediction tasks.

关键词： spatio-temporal predictive learning precipitation nowcasting neural network

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

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：