检索结果-南通市图书馆

Metarelation2vec:A Metapath-Free Scalable Representation Learning Model for Heterogeneous Networks

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Tsinghua Science and Technology 2024年第2期29卷 553-575页

作者： lei Chen Yuan Li Yong lei Xingye Deng School of Information and Electrical Engineering Hunan University of Science and TechnologyXiangtan 411201China School of Computer Science and Engineering Hunan University of Science and TechnologyXiangtan 411201China

Metapaths with specific complex semantics are critical to learning diverse semantic and structural information of heterogeneous networks(HNs)for most of the existing representation learning ***,any metapaths consisting of multiple,simple metarelations must be driven by domain *** sensitive,expensive,and limited metapaths severely reduce the flexibility and scalability of the existing models.A metapath-free,scalable representation learning model,called Metarelation2vec,is proposed for HNs with biased joint learning of all metarelations in a bid to address this ***,a metarelation-aware,biased walk strategy is first designed to obtain better training samples by using autogenerating cooperation probabilities for all metarelations rather than using expert-given ***,grouped nodes by the type,a common and shallow skip-gram model is used to separately learn structural proximity for each node ***,grouped links by the type,a novel and shallow model is used to separately learn the semantic proximity for each link ***,supervised by the cooperation probabilities of all meta-words,the biased training samples are thrown into the shallow models to jointly learn the structural and semantic information in the HNs,ensuring the accuracy and scalability of the *** experimental results on three tasks and four open datasets demonstrate the advantages of our proposed model.

关键词： metarelation random walk heterogeneous network metapath representation learning

Structural energy storage composites based on modified carbon fiber electrode with metal-organic frame enhancing layered double hydroxide

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Nano Research 2024年第3期17卷 1552-1563页

作者： Jinrui Ye Zhongbao Wang Qin lei lei Sun Jinfeng Gu School of Aerospace Engineering Beijing Institute of TechnologyBeijing 100081China Tangshan Research Institute Beijing Institute of TechnologyTangshan 063000China School of Materials Science and Engineering Shenyang Jianzhu UniversityShenyang 110168China Department of Stomatology PLA Strategic Support Force Characteristic Medical CenterBeijing 100101China Beijing Electromechanical Engineering Research Institute Beijing 100074China

Structural energy storage composites present advantages in simultaneously achieving structural strength and electrochemical *** of carbon fiber electrodes and resin structural electrolytes in energy storage composite poses challenges in maintaining good mechanical and electrochemical properties at reasonable cost and ***,we report a simple method to fabricate structural supercapacitor using carbon fiber electrodes(modified by Ni-layered double hydroxide(Ni-LDH)and in-situ growth of Co-metal-organic framework(Co-MOF)in a two-step process denoted as Co-MOF/Ni-LDH@CF)and bicontinuous-phase epoxy resin-based structural ***-MOF/Ni-LDH@CF as electrode material exhibits improved specific capacity(42.45 F·g^(-1))and cycle performance(93.3%capacity retention after 1000 cycles)in a three-electrode *** bicontinuous-phase epoxy resin-based structural electrolyte exhibits an ionic conductivity of 3.27×10^(-4) S·cm^(-1).The fabricated Co-MOF/Ni-LDH@CF/SPE-50 structural supercapacitor has an energy density of 3.21 Wh·kg^(-1) at a power density of 42.25 W·kg^(-1),whilst maintaining tensile strength and modulus of 334.6 MPa and 25.2 *** results show practical potential of employing modified commercial carbon fiber electrodes and epoxy resin-based structural electrolytes in structural energy storage applications.

关键词： carbon fiber electrode structural electrolyte structural energy storage composites supercapacitor

DCVAE-adv:A Universal Adversarial Example Generation Method for White and Black Box Attacks

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Tsinghua Science and Technology 2024年第2期29卷 430-446页

作者： lei Xu Junhai Zhai College of Mathematics and Information Science Hebei UniversityBaoding 071002China

Deep neural network(DNN)has strong representation learning ability,but it is vulnerable and easy to be fooled by adversarial *** order to handle the vulnerability of DNN,many methods have been *** general idea of existing methods is to reduce the chance of DNN models being fooled by observing some designed adversarial examples,which are generated by adding perturbations to the original *** this paper,we propose a novel adversarial example generation method,called *** from the existing methods,DCVAE-adv constructs adversarial examples by mixing both explicit and implicit perturbations without using original ***,the proposed method can be applied to both white box and black box *** addition,in the inference stage,the adversarial examples can be generated without loading the original images into memory,which greatly reduces the memory *** compared DCVAE-adv with three most advanced adversarial attack algorithms:FGSM,AdvGAN,and AdvGAN++.The experimental results demonstrate that DCVAE-adv is superior to these state-of-the-art methods in terms of attack success rate and transfer ability for targeted *** code is available at https://***/xzforeverlove/DCVAE-adv.

关键词： deep neural network adversarial examples white box attack black box attack robustness

Dynamic multifunctional metasurfaces:an inverse design deep learning approach

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Photonics Research 2024年第1期12卷 123-133页

作者： ZHI-DAN lei YI-DUO XU CHENG lei YAN ZHAO DU WANG The Institute of Technological Sciences Wuhan UniversityWuhan 430072China College of Materials Science and Engineering Sichuan UniversityChengdu 610065China

Optical metasurfaces(OMs)offer unprecedented control over electromagnetic waves,enabling advanced optical *** emergence of deep learning has opened new avenues for designing ***,existing deep learning methods for OMs primarily focus on forward design,which limits their design capabilities,lacks global optimization,and relies on prior ***,most OMs are static,with fixed functionalities once *** overcome these limitations,we propose an inverse design deep learning method for dynamic *** approach comprises a forward prediction network and an inverse retrieval *** forward prediction network establishes a mapping between meta-unit structure parameters and reflectance *** inverse retrieval network generates a library of meta-unit structure parameters based on target requirements,enabling end-to-end design of *** incorporating the dynamic tunability of the phase change material Sb2Te3with inverse design deep learning,we achieve the design and verification of dynamic multifunctional *** results demonstrate OMs with multiple information channels and encryption capabilities that can realize multiple physical field optical modulation *** Sb2Te3is in the amorphous state,near-field nano-printing based on meta-unit amplitude modulation is achieved for X-polarized incident light,while holographic imaging based on meta-unit phase modulation is realized for circularly polarized *** the crystalline state,the encrypted information remains secure even with the correct polarization input,achieving double *** research points towards ultra-compact,high-capacity,and highly secure information storage approaches.

关键词： polarization inverse approach

Two-Stage Online Debiased Lasso Estimation and Inference for High-Dimensional Quantile Regression with Streaming Data

在线全文

学校读者我要写书评

暂无评论

Journal of Systems Science & Complexity 2024年第3期37卷 1251-1270页

作者： PENG Yanjin WANG lei School of Statistics and Data Science KLMDASRLEBPS and LPMCNankai UniversityTianjin 300071China

In this paper,the authors propose a two-stage online debiased lasso estimation and statistical inference method for high-dimensional quantile regression(QR)models in the presence of streaming *** the first stage,the authors modify the QR score function based on kernel smoothing and obtain the online lasso smoothed QR estimator through iterative *** estimation process only involves the current data batch and specific historical summary statistics,which perfectly accommodates to the special structure of streaming *** the second stage,an online debiasing procedure is carried out to eliminate biases caused by the lasso penalty as well as the accumulative approximation error so that the asymptotic normality of the resulting estimator can be *** authors conduct extensive numerical experiments to evaluate the performance of the proposed *** experiments demonstrate the effectiveness of the proposed method and support the theoretical *** application to the Beijing PM2.5 Dataset is also presented.

关键词： Adaptive tuning asymptotic normality debiased lasso online updating quantile regres-sion

同方期刊数据库

The Short-Duration Heavy Rainfall in Different Quadrants of Northeast China Cold Vortices

在线全文

学校读者我要写书评

暂无评论

Journal of Meteorological Research 2024年第2期38卷 321-338页

作者： lei YANG Yongguang ZHENG Liaoning Meteorological Disaster Monitoring and Early Warning Center Shenyang 110166 National Meteorological Centre China Meteorological AdministrationBeijing 100081 China Meteorological Administration Tornado Key Laboratory Foshan 528315

The Northeast China cold vortex(NCCV)is one of the main synoptic-scale systems causing short-duration heavy rainfall(SDHR)in Northeast *** conditions(e.g.,water vapor,instability,and vertical wind shear)are known to be distinctly different over the four quadrants of NCCVs,rendering prediction of the SDHR related to NCCVs(NCCV_SDHR)more *** on 5-yr hourly rainfall observations from 3196 automatic weather stations and ERA5 reanalysis data,10,232 NCCV_SDHR events were identified and divided into four quadrant groups according to their relative position to the center of the NCCV(CVC).The results show that the southeast quadrant features the highest frequency of SDHR,with stronger intensity,longer duration,and wider coverage;and the SDHR in different quadrants presents different formation mechanisms and varied temporal evolution.A new coordinate system is established relative to the CVC that uses the CVC as the origin and the radius of the NCCV(r CV)as the unit *** this new coordinate system,all of the NCCV_SDHR events in the 5-yr study period are *** is found that the occurrence frequency of NCCV_SDHR initially increases and then decreases with increasing distance from the *** highest frequency occurs mainly between 0.8 and 2.5 times r CV from the CVC in the southeast *** can be attributed to the favorable conditions,such as convergence of the low-level shear line and abundant water vapor,which are concentrated in this ***,high-frequency NCCV_SDHR larger than 50 mm(NCCV_SDHR50)is observed to be closer to the *** NCCV_SDHR50occurs,the NCCV is in closer proximity to the subtropical high,resulting in stronger low-level convergence and more abundant water ***,there are lower lifting condensation levels and stronger 0-6-and 0-1-km vertical wind shears in these *** findings provide a valuable reference for more accurate prediction of NCCV_SDHR.

关键词： Northeast China cold vortex short-duration heavy rainfall statistical characteristics environmental conditions

Towards catalytic reactions of Cu single-atom catalysts:Recent progress and future perspective

在线全文

学校读者我要写书评

暂无评论

Chinese Chemical Letters 2024年第2期35卷 197-207页

作者： Yitao Zhao lei Tao Jiangsu Province Engineering Research Center of Special Functional Textile Materials Changzhou Vocational Institute of Textile and GarmentChangzhou 213164China Jiangsu Key Laboratory of Advanced Catalytic Materials and Technology Advanced Catalysis and Green Manufacturing Collaborative Innovation CenterSchool of Petrochemical EngineeringChangzhou UniversityChangzhou 213164China

One of the urgent and challenging topics in diversified sustainable energy conversion is the development of high-performance,low-cost,and well durable *** single-atom catalysts(SACs)have become promising catalysts for diversified sustainable energy conversion due to their capability to maximize the utilization efficiency,acquire modulated electronic structure and optimized binding strength with *** this review,we have provided an interview of the recent progress achieved in the field of electrocatalysis,photocatalysis,and heterogeneous reaction based on Cu *** by this review,we have summarized some advanced synthetic strategies for the construction of Cu ***,the performance-improving strategies are discussed in terms of the coordination environments of the reaction center,reaction mechanism and selectivity,based on free energy diagram and electron structure ***,the remaining issues,challenges,and opportunities of Cu SACs are also provided,affording a perspective for future *** review not only offers us a deep understanding on the catalytic mechanism of Cu SACs for energy conversion,but also encourages more endeavors in prompting their practical application.

关键词： Cu SACs Energy conversion Catalysis Coordination environment

Proteomic response of Phaeocystis globosa to nitrogen limitation

在线全文

学校读者我要写书评

暂无评论

Journal of Oceanology and Limnology 2024年第1期42卷 141-149页

作者： Haisu LIU Ruiwang WEI Qiangyong lei lei CUI Songhui LÜ Key Laboratory of Eutrophication and Red Tide Prevention Research Center for Harmful Algae and Marine BiologyJinan UniversityGuangzhou 510632China Southern Marine Science and Engineering Guangdong Laboratory Zhuhai 519082China National Engineering Laboratory of Electronic and Information Standardization Shenzhen 518057China

Phaeocystis globosa is an important unicellular eukaryotic alga that can also form *** can cause massive harmful algal blooms and plays an important role in the global carbon or sulfur *** far,the ecophysiology of *** has been investigated by numerous ***,the proteomic response of *** to nitrogen depletion remains largely *** compared four protein preparation methods of *** for two-dimensional electrophoresis(2-DE)(Urea/Triton X-100 with trichloroacetic acid(TCA)/acetone precipitation;TCA/acetone precipitation;Radio Immuno Precipitation Assay(RIPA)with TCA/acetone precipitation;and Tris buffer).Results show that the combination of RIPA with TCA/acetone precipitation had a clear gel background and showed the best protein spot separation effect,based on which the proteomic response to nitrogen depletion was studied using *** addition,we identified six differentially expressed proteins whose relative abundance increased or decreased more than 1.5-fold(P<0.05).Most proteins could not be identified,which might be attributed to the lack of genomic sequences of *** nitrogen limitation,replication protein-like,RNA ligase,and sn-glycerol-3-phosphate dehydrogenase were reduced,which may decrease the DNA replication level and ATP production in *** *** increase of endonucleaseⅢand transcriptional regulator enzyme may affect the metabolic and antioxidant function of *** cells and induce cell *** findings provide a basis for further proteomic study of *** and the optimization of protein preparation methods of marine microalgae.

关键词： Phaeocystis globosa nitrogen limitation proteomic response two-dimensional electrophoresis

H_(∞)/Passive Synchronization of Semi-Markov Jump Neural Networks Subject to Hybrid Attacks via an Activation Function Division Approach

在线全文

学校读者我要写书评

暂无评论

Journal of Systems Science & Complexity 2024年第3期37卷 1023-1036页

作者： ZHANG Ziwei SHEN Hao SU lei School of Electrical and Information Engineering Anhui University of TechnologyMa'anshan 243002China Anhui Province Key Laboratory of Special Heavy Load Robot and School of Electrical and Information Engineering Anhui University of TechnologyMa'anshan243002China

In this work,an H_(∞)/passive-based secure synchronization control problem is investigated for continuous-time semi-Markov neural networks subject to hybrid attacks,in which hybrid attacks are the combinations of denial-of-service attacks and deception attacks,and they are described by two groups of independent Bernoulli *** this foundation,via the Lyapunov stability theory and linear matrix inequality technology,the H_(∞)/passive-based performance criteria for semi-Markov jump neural networks are ***,an activation function division approach for neural networks is adopted to further reduce the conservatism of the ***,a simulation example is provided to verify the validity and feasibility of the proposed method.

关键词： Activation function division approach deception attacks denial-of-service attacks H_(∞)/passive synchronization semi-Markov jump neural networks