检索结果-南通市图书馆

Stochastic programming based multi-arm bandit offloading strategy for internet of things

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Digital Communications and Networks 2023年第5期9卷 1200-1211页

作者： Bin Cao Tingyong Wu xiang bai State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing100876China National Key Laboratory of Science and Technology on Communications University of Electronic Science and Technology of ChinaChengdu611731China China Electrics Technology Cyber Security Co..Ldt Chengdu611731China

In order to solve the high latency of traditional cloud computing and the processing capacity limitation of Internet of Things(IoT)users,Multi-access Edge Computing(MEC)migrates computing and storage capabilities from the remote data center to the edge of network,providing users with computation services quickly and *** this paper,we investigate the impact of the randomness caused by the movement of the IoT user on decision-making for offloading,where the connection between the IoT user and the MEC servers is *** uncertainty would be the main obstacle to assign the task ***,if the assigned task cannot match well with the real connection time,a migration(connection time is not enough to process)would be *** order to address the impact of this uncertainty,we formulate the offloading decision as an optimization problem considering the transmission,computation and *** the help of Stochastic Programming(SP),we use the posteriori recourse to compensate for inaccurate ***,in heterogeneous networks,considering multiple candidate MEC servers could be selected simultaneously due to overlapping,we also introduce the Multi-Arm Bandit(MAB)theory for MEC *** extensive simulations validate the improvement and effectiveness of the proposed SP-based Multi-arm bandit Method(SMM)for offloading in terms of reward,cost,energy consumption and *** results showthat SMMcan achieve about 20%improvement compared with the traditional offloading method that does not consider the randomness,and it also outperforms the existing SP/MAB based method for offloading.

关键词： Multi-access computing Internet of things Offloading Stochastic programming Multi-arm bandit

Confidence-weighted mutual supervision on dual networks for unsupervised cross-modality image segmentation

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2023年第11期66卷 54-68页

作者： Yajie CHEN Xin YANG xiang bai School of Electronic Information and Communications Huazhong University of Science and Technology School of Artificial Intelligence and Automation Huazhong University of Science and Technology

The unsupervised cross-modality image segmentation has gained much attention. Many methods attempt to align different modalities via adversarial learning. Recently, self-training with pseudo labels for the unsupervised target modality has also been widely used and achieved very promising results. The pseudo labels are usually obtained by selecting reliable predictions whose highest predicted probability is larger than an empirically set value. Such pseudo label generation inevitably has noise and training a segmentation model using incorrect pseudo labels could yield nontrivial errors for the target modality. In this paper, we propose a confidence-weighted mutual supervision on dual networks for unsupervised cross-modality image segmentation. Specifically, we independently initialize two networks with the same architecture, and propose a novel confidence-weighted Dice loss to mutually supervise the two networks using their predicted results for unlabeled data. In this way, we make full use of all predictions of unlabeled images and leverage the prediction confidence to alleviate the negative impact of noisy pseudo labels. Extensive experiments on three widely-used unsupervised cross-modality image segmentation datasets(i.e., MM-WHS 2017, Brats 2018, and Multi-organ segmentation) demonstrate that the proposed method achieves superior performance to some state-of-the-art methods.

关键词： domain adaptation pseudo label mutual supervision cross-modality image segmentation

同方期刊数据库

Bifurcation and chaos of a 4-side fixed rectangular thin plate in electromagnetic and mechanical fields

在线全文

学校读者我要写书评

暂无评论

Journal of Zhejiang University-Science A(Applied Physics & Engineering) 2009年第1期10卷 62-71页

作者： Wei-guo ZHU xiang-zhong bai School of Civil Engineering and Mechanics Yanshan University Qinhuangdao 066004 China Department of Transportation Engineering Huaiyin Institute of Technology Huai 'an 223001 China

We studied the problem of bifurcation and chaos in a 4-side fixed rectangular thin plate in electromagnetic and me-chanical *** on the basic nonlinear electro-magneto-elastic motion equations for a rectangular thin plate and the ex-pressions of electromagnetic forces,the vibration equations are derived for the mechanical loading in a steady transverse magnetic *** the Melnikov function method,the criteria are obtained for chaos motion to exist as demonstrated by the Smale horseshoe *** vibration equations are solved numerically by a fourth-order Runge-Kutta *** bifurcation dia-gram,Lyapunov exponent diagram,displacement wave diagram,phase diagram and Poincare section diagram are obtained.

关键词： Rectangular thin plate Electromagnetic-mechanical coupling Melnikov function method Runge-Kutta method Bifurcation Chaos

Scene text detection and recognition： recent advances and future trends

在线全文

学校读者我要写书评

暂无评论

Frontiers of Computer Science 2016年第1期10卷 19-36页

作者： Yingying ZHU Cong YAO xiang bai School of Electronic Information and Communications Huazhong University of Science and Technology Wuhan 430074 China

Text, as one of the most influential inventions of humanity, has played an important role in human life, so far from ancient times. The rich and precise information embod- ied in text is very useful in a wide range of vision-based ap- plications, therefore text detection and recognition in natu- ral scenes have become important and active research topics in computer vision and document analysis. Especially in re- cent years, the community has seen a surge of research efforts and substantial progresses in these fields, though a variety of challenges （e.g. noise, blur, distortion, occlusion and varia- tion） still remain. The purposes of this survey are three-fold： 1） introduce up-to-date works, 2） identify state-of-the-art al- gorithms, and 3） predict potential research directions in the future. Moreover, this paper provides comprehensive links to publicly available resources, including benchmark datasets, source codes, and online demos. In summary, this literature review can serve as a good reference for researchers in the areas of scene text detection and recognition.

关键词： text detection text recogntion natural image,algorithms applications

Feature context learning for human parsing

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2019年第12期62卷 6-19页

作者： Tengteng HUANG Yongchao XU Song bai Yongpan WANG xiang bai School of Electronic Information and Communications Huazhong University of Science and Technology Alibaba Group

Parsing inconsistency, referring to the scatters and speckles in the parsing results as well as imprecise contours, is a long-standing problem in human parsing. It results from the fact that the pixel-wise classification loss independently considers each pixel. To address the inconsistency issue, we propose in this paper an end-to-end trainable, highly flexible and generic module called feature context module(FCM).FCM explores the correlation of adjacent pixels and aggregates the contextual information embedded in the real topology of the human body. Therefore, the feature representations are enhanced and thus quite robust in distinguishing semantically related parts. Extensive experiments are done with three different backbone models and four benchmark datasets, suggesting that FCM can be an effective and efficient plug-in to consistently improve the performance of existing algorithms without sacrificing the inference speed too much.

关键词： human parsing context learning fully convolutional networks graph convolutional network semantic segmentation

Hot Corrosion Resistance of TB8 Titanium Alloy after ECAP and Heat Treatment

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 Wuhan University of Technology(Materials Science) 2023年第6期38卷 1440-1448页

作者： LI Shuaidi XU Xiaojing bai xiang CAO Bin Institute of Advanced Manufacturing and Modern Equipment Technology Jiangsu UniversityZhenjiang 212013China

The metastableβtitanium alloy TB8(Ti-12.76Mo-2.13Nb-2.73A1-0.16Si)was used as the original material,and the secondary processing method combining equal channel angular pressing(ECAP)and heat treatment was *** the help of optical microscope(OM),scanning electron microscope(SEM)and X-ray diffractometer(XRD),the corrosion behavior of TB8 titanium alloy after different secondary processing(800℃/850℃solid solution-520℃aging,ECAP-800℃/850℃solid solution-520℃aging,and800℃/850℃solid solution-ECAP-520℃aging)was *** experimental results show that the hot corrosion products of the six samples are similar,mainly Na_(2)Si_(2)O_(5),MoS_(2),TiCl_(2),Ti(SO_(4))_(2),and *** to the grains of the TB8 titanium alloy treated by 850℃solid solution-ECAP-520℃aging are obviously refined,the surface structure is the most smooth and dense,forming a continuous Al2O3protective film,and the surface defects are the least after *** corrosion layer thickness is the lowest(102.3μm),only 36.5%-81.4%of that of other secondary processing titanium *** addition,the corrosion kinetics curves of the six materials all follow parabolic laws,and the minimum corrosion weight gain of the samples after 850℃solutionECAP-520℃aging treatment is 0.7507 mg·mm^(-2),showing better hot corrosion resistance.

关键词： TB8 ECAP heat treatment hot corrosion resistance

Special focus on deep learning for computer vision

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2020年第2期63卷 5-6页

作者： xiang bai Yanwei PANG Guofeng ZHANG Huazhong University of Science and Technology Tianjin University Zhejiang University

Deep learning has achieved great successes in many research areas. In particular, remarkable progresses have been made in the field of computer vision. This special focus, which will also appear in the next few issues... 详细信息

关键词： image Special focus on deep learning for computer vision

Voltage-driven charge-mediated fast 180 degree magnetization switching in nanoheterostructure at room temperature

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

npj Computational Materials 2017年第1期3卷 137-145页

作者： Min Yi Hongbin Zhang bai-xiang Xu Department of Materials and Earth Sciences Technische Universitat Darmstadt64287 DarmstadtGermany

Voltage-driven 180°magnetization switching without electric current provides the possibility for revolutionizing the *** demonstrated the voltage-driven charge-mediated 180°magnetization switching at room temperature by combining firstprinciples calculations and temperature-dependent magnetization dynamics *** electric field(E)-induced interface charge is found to allow a giant modulation of the magnetic anisotropy(K)of the *** K is revealed to vary linearly with respect to E and the epitaxial *** dynamics simulations using the so-obtained K show that both inplane and perpendicular 180°switching can be achieved by E *** temperature effect renders the 180°switching as probability *** analysis indicates a fast(around 4 ns)and low-error-probability 180°switching achievable at room temperature by controlling the magnitude of E and the pulse *** study inspires the rational design of miniaturized nanoscale spintronic devices where thermal fluctuation has a great impact.

关键词： magnetization charge render

Electric field induced magnetization reversal in magnet/insulator nanoheterostructure

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

International Journal of Smart and Nano Materials 2020年第3期11卷 298-309页

作者： Qihua Gong Min Yi bai-xiang Xu Institute of Materials Science Technische Universität DarmstadtDarmstadtGermany State Key Lab of Mechanics and Control of Mechanical Structures&Key Lab for Intelligent Nano Materials and Devices of Ministry of Education&College of Aerospace Engineering Nanjing University of Aeronautics and Astronautics(NUAA)NanjingChina

Electric-field control of magnetization reversal is promising for lowpower *** in a magnet/insulator nanoheterostructure which is the fundamental unit of magnetic tunneling junction in spintronics,we demonstrate the electric field induced 180°magne-tization switching through a multiscale study combining firstprinciples calculations and finite-temperature magnetization *** the model nanoheterostructure MgO/Fe/Cu with insu-lator MgO,soft nanomagnet Fe and capping layer Cu,through firstprinciples calculations we find its magnetocrystalline anisotropy linearly varying with the electric *** finite-temperature magnetization dynamics which is informed by the first-principles results,we disclose that a room-temperature 180°magnetization switching with switching probability higher than 90%is achievable by controlling the electric-field pulse and the nanoheterostructure *** 180°switching could be fast realized within 5 *** study is useful for the design of low-power,fast,and miniaturized nanoscale electric-field-controlled spintronics.

关键词： Magnetization reversal nanoheterostructure electric field spintronics multiscale simulations switching probability