检索结果-南通市图书馆

DLP:towards active defense against backdoor attacks with decoupled learning process

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Cybersecurity 2024年第1期7卷 122-134页

作者： Zonghao Ying Bin Wu State Key Laboratory of Information Security Institute of Information EngineeringChinese Academy of SciencesBeijingChina School of Cyber Security University of Chinese Academy of SciencesBeijingChina

Deep learning models are well known to be susceptible to backdoor attack,where the attacker only needs to provide a tampered dataset on which the triggers are *** trained on the dataset will passively implant the backdoor,and triggers on the input can mislead the models during *** study shows that the model shows different learning behaviors in clean and poisoned subsets during *** on this observation,we propose a general training pipeline to defend against backdoor attacks *** models can be trained from the unreli-able dataset by decoupling the learning process into three stages,i.e.,supervised learning,active unlearning,and active semi-supervised *** effectiveness of our approach has been shown in numerous experiments across various backdoor attacks and datasets.

关键词： Deep learning Backdoor attack Active defense

Digital Twin-Assisted Semi-Federated Learning Framework for Industrial Edge Intelligence

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

china Communications 2024年第5期21卷 314-329页

作者： Wu Xiongyue Tang Jianhua Marie Siew Shien-Ming Wu School of Intelligent Engineering South China University of TechnologyGuangzhou 511442China Electrical and Computer Engineering Department Carnegie Mellon UniversityPittsburghPA 15213United States

The rapid development of emerging technologies,such as edge intelligence and digital twins,have added momentum towards the development of the Industrial Internet of Things(IIo T).However,the massive amount of data generated by the IIo T,coupled with heterogeneous computation capacity across IIo T devices,and users’data privacy concerns,have posed challenges towards achieving industrial edge intelligence(IEI).To achieve IEI,in this paper,we propose a semi-federated learning framework where a portion of the data with higher privacy is kept locally and a portion of the less private data can be potentially uploaded to the edge *** addition,we leverage digital twins to overcome the problem of computation capacity heterogeneity of IIo T devices through the mapping of physical *** formulate a synchronization latency minimization problem which jointly optimizes edge association and the proportion of uploaded nonprivate *** the joint problem is NP-hard and combinatorial and taking into account the reality of largescale device training,we develop a multi-agent hybrid action deep reinforcement learning(DRL)algorithm to find the optimal *** results show that our proposed DRL algorithm can reduce latency and have a better convergence performance for semi-federated learning compared to benchmark algorithms.

关键词： digital twin edge association industrial edge intelligence(IEI) semi-federated learning

A Reverse Path Planning Approach for Enhanced Performance of Multi-Degree-of-Freedom Industrial Manipulators

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

Computer Modeling in Engineering & sciences 2024年第5期139卷 1357-1379页

作者： Zhiwei Lin Hui Wang Tianding Chen Yingtao Jiang Jianmei Jiang Yingpin Chen Key Laboratory of Light Field Manipulation and System Integration Applications in Fujian Province School of Physics and Information EngineeringMinnan Normal UniversityZhangzhou363000China Department of Electrical and Computer Engineering University of NevadaLas Vegas89154USA College of Material Engineering Fujian Agriculture and Forestry UniversityFuzhou350000China

In the domain of autonomous industrial manipulators,precise positioning and appropriate posture selection in path planning are pivotal for tasks involving obstacle avoidance,such as handling,heat sealing,and *** Multi-Degree-of-Freedom(MDOF)manipulators offer kinematic redundancy,aiding in the derivation of optimal inverse kinematic solutions to meet position and posture requisites,their path planning entails intricate multiobjective optimization,encompassing path,posture,and joint motion *** satisfactory results in practical scenarios remains *** response,this study introduces a novel Reverse Path Planning(RPP)methodology tailored for industrial *** approach commences by conceptualizing the manipulator’s end-effector as an agent within a reinforcement learning(RL)framework,wherein the state space,action set,and reward function are precisely defined to expedite the search for an initial collision-free *** enhance convergence speed,the Q-learning algorithm in RL is augmented with ***,we formulate the cylindrical bounding box of the manipulator based on its Denavit-Hartenberg(DH)parameters and propose a swift collision detection ***,the motion performance of the end-effector is refined through a bidirectional search,and joint weighting coefficients are introduced to mitigate motion in high-power *** efficacy of the proposed RPP methodology is rigorously examined through extensive simulations conducted on a six-degree-of-freedom(6-DOF)manipulator encountering two distinct obstacle configurations and target *** results substantiate that the RPP method adeptly orchestrates the computation of the shortest collision-free path while adhering to specific posture constraints at the target ***,itminimizes both posture angle deviations and joint motion,showcasing its prowess in enhancing the operational performance of MDOF industrial manipulators.

关键词： Reverse path planning Dyna-Q bidirectional search posture angle joint motion

Numerical Analysis on the Static Performance of Gas Journal Bearing by Using Finite Element Method

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Nanomanufacturing and Metrology 2024年第1期7卷 28-47页

作者： Pengfeng Wang Yuntang Li Xiang Gao Yueliang Ye Ruirui Li Xiaolu Li Yuan Chen Jie Jin Cong Zhang College of Modern science and Technology China Jiliang UniversityJinhua 322002China College of Mechanical and Electrical Engineering China Jiliang UniversityHangzhou 310018China

In this paper,finite element method is used to calculate the static performance of gas journal bearing,in which rotation speed term is introduced into the stiffness matrix of linear triangular element to realize the performance calculation of the bearing with rotation *** results indicate that the average gas film thicknesses corresponding to the maximum load capac-ity and stiffness,and the minimum attitude angle increase with the growth of orifice *** capacity and stiffness significantly improved with the increase of rotation speed,eccentricity ratio and supply pressure when the bearing has thin average gas film *** angle increases with the growth of rotation speed,while the growth rate slows down or even decreases at high *** most effective way of reducing attitude angle is to increase supply *** can be found that rotation speed affects attitude angle through changing gas pressure difference between two orifices,while other parameters have the same effect by changing gas pressure at orifice outlet.

关键词： Finite element method(FEM) Gas journal bearing Stiffness matrix Attitude angle

Multi-modal knowledge graph inference via media convergence and logic rule

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

CAAI Transactions on Intelligence Technology 2024年第1期9卷 211-221页

作者： Feng Lin Dongmei Li Wenbin Zhang Dongsheng Shi Yuanzhou Jiao Qianzhong Chen Yiying Lin Wentao Zhu School of Information science and Technology Beijing Forestry UniversityBeijingChina Engineering Research Center for Forestry-oriented Intelligent Information Processing National Forestry and Grassland AdministrationBeijingChina Michigan Technological University HoughtonMichiganUSA

Media convergence works by processing information from different modalities and applying them to different *** is difficult for the conventional knowledge graph to utilise multi-media features because the introduction of a large amount of information from other modalities reduces the effectiveness of representation learning and makes knowledge graph inference less *** address the issue,an inference method based on Media Convergence and Rule-guided Joint Inference model(MCRJI)has been *** authors not only converge multi-media features of entities but also introduce logic rules to improve the accuracy and interpretability of link ***,a multi-headed self-attention approach is used to obtain the attention of different media features of entities during semantic ***,logic rules of different lengths are mined from knowledge graph to learn new entity ***,knowledge graph inference is performed based on representing entities that converge multi-media *** experimental results show that MCRJI outperforms other advanced baselines in using multi-media features and knowledge graph inference,demonstrating that MCRJI provides an excellent approach for knowledge graph inference with converged multi-media features.

关键词： logic rule media convergence multi-modal knowledge graph inference representation learning

A Simulation-Extrapolation Approach to the Analysis of Interval-Censored Failure Time Data with Mis-Measured Covariates

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Systems science & Complexity 2024年

作者： FENG Fan ZHAO Shishun LI Shuwei SUN Jianguo School of Economics and Statistics Guangzhou University School of Mathematics Jilin University School of Mathematics Jilin University Department of Statistics University of Missouri

Interval-censored failure time data arise frequently in periodical follow-up studies including clinical trials and epidemiological surveys. In addition, some covariates may be subject to measurement errors due to the instrumental contamination, biological variation or other reasons. For the analysis of interval-censored data with mis-measured covariates, the existing methods either assume a parametric model or rely on the availability of replicated surrogate measurements for the error-prone covariate, which both have obvious limitations. To overcome these shortcomings, the authors propose a simulation-extrapolation estimation procedure under a general class of transformation models. The resulting estimators are shown to be consistent and asymptotically normal. The numerical results obtained from a simulation study indicate that the proposed method performs reasonably well in *** particular, the proposed method can reduce the estimation bias given by the naive method that does not take measurement errors into account. Finally, the proposed method is applied to a real data set on hypobaric decompression sickness.

关键词：

Development status and prospect of underground thermal energy storage technology

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

Journal of Groundwater science and Engineering 2024年第1期12卷 92-108页

作者： Ying-nan Zhang Yan-guang Liu Kai Bian Guo-qiang Zhou Xin Wang Mei-hua Wei School of Earth science and Engineering Hebei University of EngineeringHandan 056038Hebei ProvinceChina Institute of Hydrogeology and Environmental Geology Chinese Academy of Geological SciencesShijiazhuang 050061China Technology Innovation Center of Geothermal&Hot Dry Rock Exploration and Development Ministry of Natural ResourcesShijiazhuang 050061China Shanxi Key Laboratory for Exploration and Exploitation of Geothermal Resources Taiyuan 030024China Shanxi Geological Engineering Exploration Institute Taiyuan 030024China

underground Thermal Energy Storage(UTES)store unstable and non-continuous energy underground,releasing stable heat energy on *** effectively improve energy utilization and optimize energy *** UTES technology advances,accommodating greater depth,higher temperature and multi-energy complementarity,new research challenges *** paper comprehensively provides a systematic summary of the current research status of *** categorized different types of UTES systems,analyzes the applicability of key technologies of UTES,and evaluate their economic and environmental ***,this paper identifies existing issues with UTES,such as injection blockage,wellbore scaling and corrosion,seepage and heat transfer in cracks,*** suggests deepening the research on blockage formation mechanism and plugging prevention technology,improving the study of anticorrosive materials and water treatment technology,and enhancing the investigation of reservoir fracture network characterization technology and seepage heat *** recommendations serve as valuable references for promoting the high-quality development of UTES.

关键词： Aquifer thermal energy storage Borehole thermal energy storage Cavern thermal energy storage Thermal energy storage technology Benefit evaluation

Damaged apple detection with a hybrid YOLOv3 algorithm

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Information Processing in Agriculture 2024年第2期11卷 163-171页

作者： Meng Zhang Huazhao Liang Zhongju Wang Long Wang Chao Huang Xiong Luo Schoolof Computer and Communication Engineering University of Science and Technoogy BejingBejing 1083China lShunde Innovation SchoolUniversity of Science and Technology BeijingFoshan 528399China

This paper proposes an improved You Only Look Once(YOLOv3)algorithm for automatically detecting damaged apples to promote the automation of the fruit processing *** the proposed method,a clustering method based on Rao-1 algorithm is introduced to optimize anchor box *** clustering method uses the intersection over the union to form the objective function and the most representative anchor boxes are generated for normal and damaged apple *** verify the feasibility and effectiveness of the proposed method,real apple images collected from the Internet are *** with the generic YOLOv3 and Fast Region-based Convolutional Neural Network(Fast R-CNN)algorithms,the proposed method yields the highest mean average precision value for the test ***,it is practical to apply the proposed method for intelligent apple detection and classification tasks.

关键词： Rao algorithm Apple detection Clustering Smart agriculture

Two Improved Algorithms to Compute the Minimal Bases of Univariate Matrices

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Systems science & Complexity 2024年

作者： CHEN Lingfan YAO Shanshan School of Mathematics University of Science and Technology of China School of science Beijing University of Posts and Telecommunications Key Laboratory of Mathematics and Information Networks Beijing University of Posts and Telecommunications Ministry of Education

The minimal basis of a univariate polynomial matrix M（s）∈K[s]m×nis a basis of the syzygies of the polynomial matrix M（s） with lowest possible degree,where K[s] is the univariate polynomial ring over the field of *** provides an efficient tool to compute the moving planes and moving quadratics of a rational parametric surface,which are employed to implicitize the parametric surface as a powerful implicitization *** this paper,the authors develop two improved algorithms for computing the minimal bases of polynomial *** algorithms are based on efficient methods to reduce the degrees of a set of univariate polynomial *** is shown that the computational complexities of the two algorithms are O(m2n3d2+d2n5-（2 mn4d2-1/6 m3nd）),and O(m2nd2+（n-m）n3d2+m2n2d2/n-m) respectively,where m,n are the sizes of the polynomial matrix M（s）and d is the degree of each entry of the *** new algorithms are faster than the state-of-the-art methods by experimental *** properties about the degree of the minimal basis are also provided.

关键词：