检索结果-内蒙古大学图书馆

Graph entropy guided Node Embedding Dimension Selection for Graph Neural Networks

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Luo, Gongxu Li, Jianxin Su, Jianlin Peng, Hao Yang, Carl Sun, Lichao Yu, Philip S. He, Lifang Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University China School of Computer Science and Engineering Beihang University China Shenzhen Zhuiyi Technology Co. Ltd China Department of Computer Science Emory University United States Department of Computer Science and Engineering Lehigh University United States Department of Computer Science University of Illinois Chicago United States

Graph representation learning has achieved great success in many areas, including e-commerce, chemistry, biology, etc. However, the fundamental problem of choosing the appropriate dimension of node embedding for a given graph still remains unsolved. The commonly used strategies for Node Embedding Dimension Selection (NEDS) based on grid search or empirical knowledge suffer from heavy computation and poor model performance. In this paper, we revisit NEDS from the perspective of minimum entropy principle. Subsequently, we propose a novel Minimum Graph Entropy (MinGE) algorithm for NEDS with graph data. To be specific, MinGE considers both feature entropy and structure entropy on graphs, which are carefully designed according to the characteristics of the rich information in them. The feature entropy, which assumes the embeddings of adjacent nodes to be more similar, connects node features and link topology on graphs. The structure entropy takes the normalized degree as basic unit to further measure the higher-order structure of graphs. Based on them, we design MinGE to directly calculate the ideal node embedding dimension for any graph. Finally, comprehensive experiments with popular Graph Neural Networks (GNNs) on benchmark datasets demonstrate the effectiveness and generalizability of our proposed MinGE. Copyright © 2021, The Authors. All rights reserved.

关键词： Graph neural networks

A novel data glove design with parallelogram structure based on linkage mechanism of human hand

学校读者我要写书评

暂无评论

A novel data glove design with parallelogram structure based...

CSAA/IET International Conference on Aircraft Utility Systems 2020, AUS 2020

作者： Chen, Zhuoyuan Zhou, Jianfeng Wang, Shaoping Zhang, Pengyu Zhang, Chao Huang, Shouqing School of Automation Science and Electrical Engineering Beihang University Beijing China Beijing Advanced Innovation Center for Big-Data Based Precision Medicine Beihang University Beijing China The Department of Mechanical and Aerospace Engineering Case Western Reseerve University Cleveland United States Beijing Institute of Spacecraft Environment Engineering China Academy of Space Technology Beijing China

ISBN: (纸本)9781839535635

In this paper, a noval type of data glove is designed based on the principle of parallelogram, which aims to detect gestures in real time. In addition to the flexion and extension angles, this glove can detect the adduction and abduction angles of each finger. Because the overall structure of this data glove is optimized according to the linkage mechanism among joints of human hand. The data glove has the advantages of simple structure, convenient wearing, light weight and good comfort. The glove is composed of four parts: framework, bearings, connecting rods and potentiometers. The framework and connecting rods are respectively made of photosensitive resin and laser cut carbon fiber board. Potentiometers are installed on the bearings to detect the angle of rotation. In this paper, kinematic modeling of human hand is carried out. When knowing the rotation angles of each joint of human hand, the posture of the hand can be calculated by the model. Finally, the data glove is tested on the wearer's hand. Angles of each joint measured by the data glove is compared with the actual angles to verify the accuracy and feasibility of this design. © 2020 Institution of engineering and Technology. All rights reserved.

关键词： Structural design

Wireless Acoustic Sensor Networks and Edge Computing for Rapid Acoustic Monitoring

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2019年第1期6卷 64-74页

作者： Zhengguo Sheng Saskia Pfersich Alice Eldridge Jianshan Zhou Daxin Tian Victor C.M.Leung the Department of Engineering and Design University of Sussex Infineon Technologies AG the Department of Music University of Sussex Beijing Advanced Innovation Center for Big Data and Brain Computing (BDBC) Beijing Key Laboratory for Cooperative Vehicle Infrastructure Systems & Safety ControlSchool of Transportation Science and EngineeringBeihang University the Department of Electrical and Computer Engineering The University of British Columbia

Passive acoustic monitoring is emerging as a promising solution to the urgent, global need for new biodiversity assessment methods. The ecological relevance of the soundscape is increasingly recognised, and the affordability of robust hardware for remote audio recording is stimulating international interest in the potential for acoustic methods for biodiversity *** scale of the data involved requires automated methods,however, the development of acoustic sensor networks capable of sampling the soundscape across time and space and relaying the data to an accessible storage location remains a significant technical challenge, with power management at its core. Recording and transmitting large quantities of audio data is power intensive,hampering long-term deployment in remote, off-grid locations of key ecological interest. Rather than transmitting heavy audio data, in this paper, we propose a low-cost and energy efficient wireless acoustic sensor network integrated with edge computing structure for remote acoustic monitoring and in situ *** and computation of acoustic indices are carried out directly on edge devices built from low noise primo condenser microphones and Teensy microcontrollers, using internal FFT hardware support. Resultant indices are transmitted over a ZigBee-based wireless mesh network to a destination *** tests of audio quality, indices computation and power consumption demonstrate acoustic equivalence and significant power savings over current solutions.

关键词： Acoustic sensor networks edge computing energy efficiency

Single-shot super-resolution imaging via discernibility in the high-dimensional light-field space based on ghost imaging

学校读者我要写书评

暂无评论

Photonics Research 2025年第6期13卷 1709-1725页

作者： Zhishen Tong Chenyu Hu Jian Wang Youheng Zhu Xia Shen Zhentao Liu Shensheng Han Aerospace Laser Technology and System Department Shanghai Institute of Optics and Fine Mechanics Chinese Academy of Sciences Shanghai 201800 China Wangzhijiang Innovation Center for Laser Aerospace Laser Technology and System Department Shanghai Institute of Optics and Fine Mechanics Chinese Academy of Sciences Shanghai 201800 China School of Physics and Optoelectronic Engineering Hangzhou Institute for Advanced Study University of Chinese Academy of Sciences Hangzhou 310024 China School of Data Science Fudan University Shanghai 200433 China Center of Materials Science and Optoelectronics Engineering University of Chinese Academy of Sciences Beijing 100049 China e-mail: ztliu@*** e-mail: sshan@***

Super-resolution (SR) imaging has been widely used in several fields like remote sensing and microscopy. However, it is challenging for existing SR approaches to capture SR images in a single shot, especially in dynamic imaging scenarios. In this study, we present a single-shot SR imaging scheme that leverages discernibility in the high-dimensional (H-D) light-field space based on a ghost imaging camera via sparsity constraints (GISC camera), which is capable of encoding H-D imaging information into a two-dimensional speckle pattern detected in a single shot. We demonstrate both theoretically and experimentally that while the resolution in the H-D light-field space, characterized by the second-order light-field correlation, remains limited by light-field diffraction, the single-shot spatial resolution is greatly improved beyond classical Rayleigh’s criterion by utilizing the discernibility in the H-D light-field space. We further quantify the effects of the sampling number, signal-to-noise ratio, and object sparsity on the resolution. Our results offer significant potential for the SR observation of high-speed dynamic processes.

关键词： Ghost imaging Imaging systems Imaging techniques Near field scanning optical microscopy Optical systems Remote sensing

Side-channel-free quantum key distribution with practical devices

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Jiang, Cong Yu, Zong-Wen Hu, Xiao-Long Wang, Xiang-Bin Jinan Institute of Quantum Technology Shandong Jinan250101 China State Key Laboratory of Low Dimensional Quantum Physics Department of Physics Tsinghua University Beijing100084 China Data Communication Science and Technology Research Institute Beijing100191 China School of Physics State Key Laboratory of Optoelectronic Materials and Technologies Sun Yat-sen University Guangzhou510275 China Shanghai Branch CAS Center for Excellence and Synergetic Innovation Center in Quantum Information and Quantum Physics University of Science and Technology of China Shanghai201315 China Shenzhen Institute for Quantum Science and Engineering Physics Department Southern University of Science and Technology Shenzhen518055 China Frontier Science Center for Quantum Information Beijing China

Based on the idea that there is no side channel in the vacuum state, the side-channel-free quantum key distribution (SCFQKD) protocol was proposed, which is immune to all attacks in the source side-channel space and all attacks in the detectors. In the original SCFQKD protocol, an important assumption is that Alice and Bob can produce the perfect vacuum pulses. But due to the finite extinction ratio of the intensity modulators, the perfect vacuum pulse is impossible in practice. In this paper, we solve this problem and make the quantum key distribution side-channel secure with real source device which does not emit perfect vacuum pulses. Our conclusion only depends on the upper bounds of the intensities of the sources. No other assumptions such as stable sources and stable side channels are needed. The numerical results show that, comparing with the results of SCFQKD protocol with perfect vacuum sources, the key rates and secure distance are only slightly decreased if the upper bound of the intensity of the imperfect vacuum source is less than 10−8 which can be achieved in experiment by two-stage intensity modulator. We also show that the two-way classical communication can be used to the data post-processing of SCFQKD protocol to improve the key rate. Specially, the active odd-parity pairing method can improve the key rates in all distances by about two times and the secure distance by about 40 km. Give that the side channel security based on imperfect vacuum, this work makes it possible to realize side channel secure QKD with real devices. Copyright © 2022, The Authors. All rights reserved.

关键词： Quantum cryptography

Brain Age Estimation from MRI Using Cascade Networks with Ranking Loss

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Cheng, Jian Liu, Ziyang Guan, Hao Wu, Zhenzhou Zhu, Haogang Jiang, Jiyang Wen, Wei Tao, Dacheng Liu, Tao Beijing Advanced Innovation Center for Big Data-Based Precision Medicine Beijing China School of Computer Science and Engineering Beihang University Beijing China Beijing Advanced Innovation Center for Biomedical Engineering The School of Biological Science and Medical Engineering Beihang University China UBTech Sydney Artificial Intelligence Institute School of Computer Science FEIT University of Sydney DarlingtonNSW Australia JD Explore Academy at *** China BioMind Technology Center Beijing TianTan Hospital Beijing China Centre for Healthy Brain Ageing School of Psychiatry University of New South Wales SydneyNSW Australia

Chronological age of healthy people is able to be predicted accurately using deep neural networks from neuroimaging data, and the predicted brain age could serve as a biomarker for detecting aging-related diseases. In this paper, a novel 3D convolutional network, called two-stage-age-network (TSAN), is proposed to estimate brain age from T1-weighted MRI data. Compared with existing methods, TSAN has the following improvements. First, TSAN uses a two-stage cascade network architecture, where the first-stage network estimates a rough brain age, then the second-stage network estimates the brain age more accurately from the discretized brain age by the first-stage network. Second, to our knowledge, TSAN is the first work to apply novel ranking losses in brain age estimation, together with the traditional mean square error (MSE) loss. Third, densely connected paths are used to combine feature maps with different scales. The experiments with 6586 MRIs showed that TSAN could provide accurate brain age estimation, yielding mean absolute error (MAE) of 2.428 and Pearson’s correlation coefficient (PCC) of 0.985, between the estimated and chronological ages. Furthermore, using the brain age gap between brain age and chronological age as a biomarker, Alzheimer’s disease (AD) and Mild Cognitive Impairment (MCI) can be distinguished from healthy control (HC) subjects by support Copyright © 2021, The Authors. All rights reserved.

关键词： Mean square error

Sending-or-Not-Sending Twin-Field Quantum Key Distribution with Redundant Space

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Xu, Hai Hu, Xiao-Long Jiang, Cong Yu, Zong-Wen Wang, Xiang-Bin State Key Laboratory of Low Dimensional Quantum Physics Department of Physics Tsinghua University Beijing100084 China Frontier Science Center for Quantum Information Beijing China School of Physics State Key Laboratory of Optoelectronic Materials and Technologies Sun Yat-sen University Guangzhou510275 China Jinan Institute of Quantum technology SAICT Jinan250101 China Data Communication Science and Technology Research Institute Beijing100191 China Shanghai Branch CAS Center for Excellence and Synergetic Innovation Center in Quantum Information and Quantum Physics University of Science and Technology of China Shanghai201315 China Shenzhen Institute for Quantum Science and Engineering Physics Department Southern University of Science and Technology Shenzhen518055 China

We propose to adopt redundant space such as polarization mode in the sending-or-not-sending Twin-Field quantum key distribution (TF-QKD) in the Fock space. With the help of redundant space such as photon polarization, we can post-select events according to the outcome of the observation to the additional quantity. This compresses the bit-flip error rate in the post-selected events of the SNS protocol. The calculation shows that the method using redundant space can greatly improve the performance in practical TF-QKD, especially when the total number of pulses is small. Copyright © 2022, The Authors. All rights reserved.

关键词： Quantum cryptography

Improving the lifestyle behavior of type 2 diabetes mellitus patients using a mobile application

学校读者我要写书评

暂无评论

Improving the lifestyle behavior of type 2 diabetes mellitus...

IEEE Symposium on Computers and Communications (ISCC)

作者： Andreia Pinto João Viana Diogo Martinho Vítor Crista José Miguel Diniz Joana Reis João Matos David Greer Goreti Marreiros Alberto Freitas Department of Community Medicine Information and Health Decision Sciences (MEDCIDS) Faculty of Medicine Center for Health Technology and Services Research (CINTESIS) University of Porto Porto Portugal Research Group on Intelligent Engineering and Computing for Advanced Innovation and Development (GECAD) Institute of Engineering Polytechnic of Porto Porto Portugal Health Data Science PhD Program Faculty of Medicine Center for Health Technology and Services Research (CINTESIS) University of Porto Lisboa Portugal CLOO - Behavioral Insights Unit Faculty of Psychology CICPSI - Research Center for Psychological Science University of Lisbon Lisboa Portugal Behavioral Insights Unit CLOO Porto Portugal Behavioral Insights Unit CLOO Lisboa Portugal

ISBN: (数字)9781665497923

ISBN: (纸本)9781665497930

Type 2 diabetes is an increasingly prevalent disease and patients do not always manage the disease properly. Therefore, creating tools that help diabetics self-manage their condition over time is of the utmost importance. Technological tools that include features meeting this population's needs, with an integrated personalized feedback system, in an environment of gamified incentives, may be the way for developing a sustained app's usage. This work aims to present a still in development, novel approach to managing diabetes type 2 with a focus on data analysis from user-inputs, gamification, and personalized coaching, with an accessible user interface. Thus, we will present some functionalities, in particular the theoretical and practical concept that is behind this development, namely the Transtheoretical model of behavior change to create different profiles, to provide customized feedback according to the user's behavioral stage, and also to ascertain if it induces any improvement at the behavioral level.

关键词： Computers data analysis Computational modeling User interfaces Diabetes Behavioral sciences Mobile applications

Intense harmonics with time-varying orbital angular momentum from relativistic plasma mirrors

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Wang, Jingwei Zepf, Matt Leng, Yuxin Li, Ruxin Rykovanov, Sergey G. Shanghai 201800 China Collaborative Innovation Center of IFSA Shanghai Jiao Tong University Shanghai 200240 China Helmholtz Institute Jena Fröbelstieg 3 Jena07743 Germany Faculty of Physics and Astronomy Friedrich-Schiller-Universität Jena Jena07743 Germany High Performance Computing and Big Data Laboratory Center for Computational and Data-Intensive Science and Engineering Skolkovo Institute of Science and Technology Moscow121205 Russia

In this Letter using three-dimensional particle-in-cell simulations and analytical considerations we demonstrate intense high-order plasma surface harmonics carrying a time-varying orbital angular momentum (OAM) - the self-torque. We show that by using two laser beams with different OAMs l1 and l2 and a certain delay between each other and shooting them obliquely on an overdense plasma target, one can generate harmonics with OAM spanning nl1 to nl2, where n is the order of the harmonic. Such intense self-torqued harmonics can offer new possibilities in ultrafast spectroscopy. © 2021, CC BY.

关键词： Harmonic analysis