检索结果-内蒙古大学图书馆

Discriminative feature encoding for intrinsic image decomposition

Computational Visual Media 2023年第3期9卷 597-618页

作者： Zongji Wang Yunfei Liu Feng Lu Key Laboratory of Network Information System Technology(NIST) Aerospace Information Research InstituteChinese Academy of SciencesBeijing 100190China State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and EngineeringBeihang UniversityBeijing 100191China Peng Cheng Laboratory Shenzhen 518000China

Intrinsic image decomposition is an important and long-standing computer vision *** an input image,recovering the physical scene properties is *** physically motivated priors have been used to restrict the solution space of the optimization problem for intrinsic image *** work takes advantage of deep learning,and shows that it can solve this challenging computer vision problem with high *** focus lies in the feature encoding phase to extract discriminative features for different intrinsic layers from an input *** achieve this goal,we explore the distinctive characteristics of different intrinsic components in the high-dimensional feature embedding *** define feature distribution divergence to efficiently separate the feature vectors of different intrinsic *** feature distributions are also constrained to fit the real ones through a feature distribution *** addition,a data refinement approach is provided to remove data inconsistency from the Sintel dataset,making it more suitable for intrinsic image *** method is also extended to intrinsic video decomposition based on pixel-wise correspondences between adjacent *** results indicate that our proposed network structure can outperform the existing state-of-the-art.

关键词： intrinsic image decomposition deep learning feature distribution data refinement

来源：评论

学校读者我要写书评

暂无评论

MMH-FE:AMulti-Precision and Multi-Sourced Heterogeneous Privacy-Preserving Neural network Training Based on Functional Encryption

引用

computers, Materials & Continua 2025年第3期82卷 5387-5405页

作者： Hao Li Kuan Shao Xin Wang Mufeng Wang Zhenyong Zhang The State Key Laboratory of Public Big Data College of Computer Science and TechnologyGuizhou UniversityGuiyang550025China Key Laboratory of Computing Power Network and Information Security Ministry of EducationShandong Computer Science CenterQilu University of Technology(Shandong Academy of Sciences)Jinan250014China China Industrial Control Systems Cyber Emergency Response Team Beijing100040China

Due to the development of cloud computing and machine learning,users can upload their data to the cloud for machine learning model ***,dishonest clouds may infer user data,resulting in user data *** schemes have achieved secure outsourced computing,but they suffer from low computational accuracy,difficult-to-handle heterogeneous distribution of data from multiple sources,and high computational cost,which result in extremely poor user experience and expensive cloud computing *** address the above problems,we propose amulti-precision,multi-sourced,andmulti-key outsourcing neural network training ***,we design a multi-precision functional encryption computation based on Euclidean ***,we design the outsourcing model training algorithm based on a multi-precision functional encryption with multi-sourced ***,we conduct experiments on three *** results indicate that our framework achieves an accuracy improvement of 6%to 30%.Additionally,it offers a memory space optimization of 1.0×2^(24) times compared to the previous best approach.

关键词： Functional encryption multi-sourced heterogeneous data privacy preservation neural networks

来源：评论

学校读者我要写书评

暂无评论

Pushing AI to wireless network edge: an overview on integrated sensing, communication, and computation towards 6G

引用

Science China(Information Sciences) 2023年第3期66卷 7-25页

作者： Guangxu ZHU Zhonghao LYU Xiang JIAO Peixi LIU Mingzhe CHEN Jie XU Shuguang CUI Ping ZHANG Shenzhen Research Institute of Big Data Future Network of Intelligence Institute (FNii) The Chinese University of Hong Kong (Shenzhen) School of Science and Engineering (SSE) The Chinese University of Hong Kong (Shenzhen) State Key Laboratory of Advanced Optical Communication Systems and Networks School of Electronics Peking University Department of Electrical and Computer Engineering and Institute for Data Science and Computing University of Miami Peng Cheng Laboratory State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications

Pushing artificial intelligence(AI) from central cloud to network edge has reached board consensus in both industry and academia for materializing the vision of artificial intelligence of things(AIoT) in the sixth-generation(6G) era. This gives rise to an emerging research area known as edge intelligence, which concerns the distillation of human-like intelligence from the vast amount of data scattered at the wireless network edge. Typically, realizing edge intelligence corresponds to the processes of sensing, communication,and computation, which are coupled ingredients for data generation, exchanging, and processing, ***, conventional wireless networks design the three mentioned ingredients separately in a task-agnostic manner, which leads to difficulties in accommodating the stringent demands of ultra-low latency, ultra-high reliability, and high capacity in emerging AI applications like auto-driving and metaverse. This thus prompts a new design paradigm of seamlessly integrated sensing, communication, and computation(ISCC) in a taskoriented manner, which comprehensively accounts for the use of the data in downstream AI tasks. In view of its growing interest, this study provides a timely overview of ISCC for edge intelligence by introducing its basic concept, design challenges, and enabling techniques, surveying the state-of-the-art advancements, and shedding light on the road ahead.

关键词： sixth-generation (6G) edge intelligence artificial intelligence of things (AIoT) integrated sensing communication and computation (ISCC)

来源：评论

学校读者我要写书评

暂无评论

Extinction Chains Reveal Intermediate Phases Between the Safety and Collapse in Mutualistic Ecosystems

引用

Engineering 2024年第12期43卷 89-98页

作者： Guangwei Wang Xueming Liu Ying Xiao Ye Yuan Linqiang Pan Xiaohong Guan Jianxi Gao Hai-Tao Zhang The MOE Engineering Research Center of Autonomous Intelligent Unmanned Systems State Key Laboratory of Digital Manufacturing Equipment and TechnologySchool of Artificial Intelligence and AutomationHuazhong University of Science and TechnologyWuhan 430074China Guangdong HUST Industrial Technology Research Institute Huazhong University of Science and TechnologyDongguan 523808China Guangdong Provincial Engineering Technology Center of Autonomous Unmanned Vessels Dongguan 523808China MOE Key Laboratory of Intelligent Networks and Network Security&State Key Laboratory of Manufacturing Systems Xi’an Jiaotong UniversityXi’an 710049China Tsinghua National Laboratory of Information Science and Technology Department of AutomationTsinghua UniversityBeijing 100084China Department of Computer Science Rensselaer Polytechnic InstituteTroyNY 12180USA

Ecosystems are undergoing unprecedented persistent deterioration due to unsustainable anthropogenic human activities,such as overfishing and deforestation,and the effects of such damage on ecological stability are *** recent advances in experimental and theoretical studies on regime shifts and tipping points,theoretical tools for understanding the extinction chain,which is the sequence of species extinctions resulting from overexploitation,are still lacking,especially for large-scale nonlinear networked *** this study,we developed a mathematical tool to predict regime shifts and extinction chains in ecosystems under multiple exploitation situations and verified it in 26 real-world mutualistic networks of various sizes and *** discovered five phases during the exploitation process:safe,partial extinction,bistable,tristable,and collapse,which enabled the optimal design of restoration strategies for degraded or collapsed *** validated our approach using a 20-year dataset from an eelgrass restoration ***,we also found a specific region in the diagram spanning exploitation rates and competition intensities,where exploiting more species helps increase *** computational tool provides insights into harvesting,fishing,exploitation,or deforestation plans while conserving or restoring the biodiversity of mutualistic ecosystems.

关键词： Complex system network science Overexploitation Regime shift Metastability

来源：评论

学校读者我要写书评

暂无评论

On a novel tracking differentiator design based on iterative learning in a moving window

引用

Control Theory and Technology 2023年第1期21卷 46-55页

作者： Xiangyang Li Rafal Madonski Zhiqiang Gao Senping Tian Key Laboratory of Autonomous Systems and Network Control Ministry of EducationSchool of Automation Science and EngineeringSouth China University of TechnologyGuangzhou 510641China Energy and Electricity Research Center Jinan UniversityZhuhai 519070China Department of Electrical and Computer Science Cleveland State UniversityClevelandOH 44115USA

Differential signals are key in control engineering as they anticipate future behavior of process variables and therefore are critical in formulating control laws such as proportional-integral-derivative(PID).The practical challenge,however,is to extract such signals from noisy measurements and this difficulty is addressed first by *** in the form of linear and nonlinear tracking differentiator(TD).While improvements were made,TD did not completely resolve the conflict between the noise sensitivity and the accuracy and timeliness of the *** two approaches proposed in this paper start with the basic linear TD,but apply iterative learning mechanism to the historical data in a moving window(MW),to form two new iterative learning tracking differentiators(IL-TD):one is a parallel IL-TD using an iterative ladder network structure which is implementable in analog circuits;the other a serial IL-TD which is implementable digitally on any computer *** algorithms are validated in simulations which show that the proposed two IL-TDs have better tracking differentiation and de-noise performance compared to the existing linear TD.

关键词： Tracking differentiator(TD) Iterative learning Iterative learning tracking differentiator(IL-TD) Active disturbance rejection control(ADRC)-Two-dimensional system(2-D system)

来源：评论

学校读者我要写书评

暂无评论

A novel deep generative model for mRNA vaccine development: Designing 5′ UTRs with N1-methyl-pseudouridine modification

引用

Acta Pharmaceutica Sinica B 2024年第4期14卷 1814-1826页

作者： Xiaoshan Tang Miaozhe Huo Yuting Chen Hai Huang Shugang Qin Jiaqi Luo Zeyi Qin Xin Jiang Yongmei Liu Xing Duan Ruohan Wang Lingxi Chen Hao Li Na Fan Zhongshan He Xi He Bairong Shen Shuai Cheng Li Xiangrong Song Institute of Systems Genetics Department of Critical Care MedicineFrontiers Science Center for Disease-related Molecular NetworkState Key Laboratory of Biotherapy and Cancer CenterWest China HospitalSichuan UniversityChengdu 610000China Department of Computer Science City University of Hong KongHong Kong 99907China Department of Biology Brandeis UniversityBostonMA 02453USA

Efficient translation mediated by the 5'untranslated region(5'UTR)is essential for the robust efficacy of mRNA ***,the N1-methyl-pseudouridine(m1)modification of mRNA can impact the translation efficiency of the 5'*** discovered that the optimal 5'UTR for m1y-modified mRNA(m1y-5'UTR)differs significantly from its unmodified counterpart,high-lighting the need for a specialized tool for designing mly-5'UTRs rather than directly utilizing high-expression endogenous gene 5'*** response,we developed a novel machine learning-based tool,Smart5UTR,which employs a deep generative model to identify superior m1y-5'UTRs in *** tailored loss function and network architecture enable Smart5UTR to overcome limitations inherent in existing *** a result,Smart5UTR can successfully design superior 5'UTRs,greatly benefiting mRNA vaccine ***,Smart5UTR-designed superior 5'UTRs significantly enhanced antibody titers induced by COVID-19 mRNA vaccines against the Delta and Omicron variants of SARS-CoV-2,surpassing the performance of vaccines using high-expression endogenous gene 5'UTRs.

关键词： mRNA vaccine Machine learning 5'UTR mRNA design Sequence design N1-Methyl-pseudouridine COVID-19 SARS-CoV-2

来源：评论

学校读者我要写书评

暂无评论

New Cryptographic Bent Functions from Permutations and Linear Translator in Cyber Security

New Cryptographic Bent Functions from Permutations and Linea...

引用

2023 IEEE International Conference on Dependable, Autonomic and Secure Computing, 2023 International Conference on Pervasive Intelligence and Computing, 2023 International Conference on Cloud and Big Data Computing, 2023 International Conference on Cyber Science and Technology Congress, DASC/PiCom/CBDCom/CyberSciTech 2023

作者： Liu, Qian Chen, Yan Chen, Zhixiong Liu, Ximeng Lin, Hui College of Computer and Data Science Fuzhou University Fuzhou China Putian University Fujian Key Laboratory of Financial Information Processing Putian China College of Computer and Data Science Fuzhou University Key Laboratory of Information Security of Network Systems Fuzhou China College of Computer and Cyber Security Fujian Normal University Fuzhou China

ISBN: (纸本)9798350304602

Cryptographic bent functions are maximally nonlinear Boolean functions with an even number of variables. They are not only closely related to some interesting combinatorial objects, but also have important applications in coding, sequence design, cryptography and cyber security. In this paper, we firstly investigate the Walsh transform of two families of functions via Maiorana-MacFarland's class. Secondly, we present new infinite families of permutations. We show that those families have nice property that one can select three elements among them which can be used to construct bent functions. Finally, by using two linear translators, we construct bent functions from the class of Maiorana-MacFarland. © 2023 IEEE.

关键词： Walsh transforms

来源：评论

学校读者我要写书评

暂无评论

Some Classes of Cryptographic Power Functions and Permutation Polynomials with Low (-1)-Differential Uniformity in Cyber Security

Some Classes of Cryptographic Power Functions and Permutatio...

引用

作者： Liu, Qian Dong, Xiaobei Chen, Zhixiong Liu, Ximeng Xu, Li College of Computer and Data Science Fuzhou University Fuzhou China Putian University Fujian Key Laboratory of Financial Information Processing Putian China College of Computer and Data Science Fuzhou University Key Laboratory of Information Security of Network Systems Fuzhou China College of Computer and Cyber Security Fujian Normal University Fuzhou China

ISBN: (纸本)9798350304602

Cryptographic functions with low differential uniformity have wide applications in cryptography and cyber security. In this paper, we further investigate c-differential uniformity proposed by Ellingsen et al. (IEEE Transactions on Information Theory. 2020, 66(9): 5781-5789). Specifically, two classes of power functions with low (1)-differential uniformity over finite fields with odd characteristic are obtained. In addition, a class of permutation polynomials and P1N or AP1N functions over F3n are proposed. These functions with low c-differential uniformity are utilized to resist against c-differential attacks in cyberspace from those with the usual low differential uniformity. © 2023 IEEE.

关键词： Cybersecurity

来源：评论

学校读者我要写书评

暂无评论

A data based system representation

引用

IFAC-PapersOnLine 2023年第2期56卷 5518-5524页

作者： Zoltán Szabó József Bokor Péter Gáspár Systems and Control Laboratory Institute for Computer Science and Control (SZTAKI) Eötvös Loránd Research Network (ELKH) Budapest Hungary Systems and Control Laboratory Institute for Computer Science and Control (SZTAKI) Eotvos Loránd Research Network (ELKH) Budapest Hungary

The paper proposes a system representation formed by a minimal collection of sufficiently long restricted trajectories generated by an observable discrete time LTI system. Conditions are given under which such a collection is a system representation and also an exhaustive parametrization of these representations is provided. These can be also interpreted as a generalized persistency condition which complements the results encountered for the controllable case. In terms of the proposed representation some system properties are investigated and a controllable-autonomous decomposition is given. Finally it is shown how the representation associated to the inverse system, to the parallel and cascade connection, respectively, can be derived.

关键词： Modeling LTI system representation data based

来源：评论

学校读者我要写书评

暂无评论

Sea-Land Segmentation of Remote Sensing Images Based on SDW-UNet

引用

computer systems Science & Engineering 2023年第5期45卷 1033-1045页

作者： Tianyu Liu Pengyu Liu Xiaowei Jia Shanji Chen Ying Ma Qian Gao The Information Department Beijing University of TechnologyBeijing 100124China School of Physics and Electronic Information Engineering Qinghai Minzu UniversityXining810000China Advanced Information Network Beijing Laboratory Beijing100124China Computational Intelligence and Intelligent Systems Beijing key Laboratory Beijing100124China Department of Computer Science University of Pittsburgh15260USA

Image segmentation of sea-land remote sensing images is of great importance for downstream applications including shoreline extraction,the monitoring of near-shore marine environment,and near-shore target *** mitigate large number of parameters and improve the segmentation accuracy,we propose a new Squeeze-Depth-Wise UNet(SDW-UNet)deep learning model for sea-land remote sensing image *** proposed SDW-UNet model leverages the squeeze-excitation and depth-wise separable convolution to construct new convolution modules,which enhance the model capacity in combining multiple channels and reduces the model *** further explore the effect of position-encoded information in NLP(Natural Language Processing)domain on sea-land segmentation *** have conducted extensive experiments to compare the proposed network with the mainstream segmentation network in terms of accuracy,the number of parameters and the time cost for *** test results on remote sensing data sets of Guam,Okinawa,Taiwan China,San Diego,and Diego Garcia demonstrate the effectiveness of SDW-UNet in recognizing different types of sea-land areas with a smaller number of parameters,reduces prediction time cost and improves performance over other mainstream segmentation *** also show that the position encoding can further improve the accuracy of model segmentation.

关键词： Sea-land segmentation UNet depth-wise separable convolution squeeze-excitation position encoding

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：