检索结果-内蒙古大学图书馆

Rubik: A hierarchical architecture for efficient graph learning

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Chen, Xiaobing Wang, Yuke Xie, Xinfeng Hu, Xing Basak, Abanti Liang, Ling Yan, Mingyu Deng, Lei Ding, Yufei Du, Zidong Chen, Yunji Xie, Yuan State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences University of Chinese Academy of Sciences Beijing100190 China State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China Department of Computer Science University of California Santa Barbara United States Department of Electrical and Computer Engineering University of California Santa Barbara United States

Graph convolutional network (GCN) emerges as a promising direction to learn the inductive representation in graph data commonly used in widespread applications, such as E-commerce, social networks, and knowledge graphs. However, learning from graphs is non-trivial because of its mixed computation model involving both graph analytics and neural network computing. To this end, we decompose the GCN learning into two hierarchical paradigms: graph-level and node-level computing. Such a hierarchical paradigm facilitates the software and hardware accelerations for GCN learning. We propose a lightweight graph reordering methodology, incorporated with a GCN accelerator architecture that equips a customized cache design to fully utilize the graph-level data reuse. We also propose a mapping methodology aware of data reuse and task-level parallelism to handle various graphs inputs effectively. Results show that Rubik accelerator design improves energy efficiency by 26.3x to 1375.2x than GPU platforms across different datasets and GCN models. Copyright © 2020, The Authors. All rights reserved.

关键词： Graph neural networks

A testbed for investigation of selective laser melting at elevated atmospheric pressure

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Griggs, David A. Gibbs, Jonathan S. Baker, Stuart P. Penny, Ryan W. Feldmann, Martin C. Hart, A. John Department of Mechanical Engineering Massachusetts Institute of Technology CambridgeMA02139 United States Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology CambridgeMA02139 United States Department of Naval Architecture and Ocean Engineering U.S. Naval Academy AnnapolisMD United States Materials and Manufacturing Directorate Air Force Research Laboratory DaytonOH United States

Metal additive manufacturing (AM) by laser powder bed fusion (L-PBF) builds upon fundamentals established in the field of laser welding which include the influence of gas and plume dynamics on weld depth and quality. L-PBF demands a thorough investigation of the complex thermophysical phenomena that occur where the laser interacts with the metal powder bed. In particular, melt pool turbulence and evaporation are influenced by the ambient gas chemistry and pressure. This paper presents the design and validation of high pressure laser melting (HPLM) testbed;this accommodates bare metal plate samples as well as manually-coated single powder layers, and operates at up to 300 psig. The open architecture of this testbed allows for full control of all relevant laser parameters in addition to ambient gas pressure and gas flow over the build area. Representative melt tracks and rasters on bare plate and powder are examined in order to validate system performance, and preliminary analysis concludes that pressure has a significant impact on melt pool aspect ratio. The HPLM system thus enables careful study pressure effects on processing of common L-PBF materials, and can be applied in the future to materials that are challenging to process under ambient pressure, such as those with high vapor pressures. © 2021, CC BY.

关键词： Lakes

The Mercedes water Cherenkov detector

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Assis, P. Bakalová, A. Barres De Almeida, U. Brogueira, P. Conceição, R. De Angelis, A. Gibilisco, L. González, B.S. Guillén, A. La Mura, G. Mendes, L.M.D. Mendes, L.F. Pimenta, M. Shellard, R.C. Tomé, B. Vícha, J. Laboratório de Instrumentac ão e Física Experimental de Partículas Lisbon Portugal Universidade de Lisboa Lisbon Portugal Institute of Physics The Czech Academy of Sciences Prague Czech Republic Rio de Janeiro Brazil Università di Padova INFN INAF PadovaI-35131 Italy Computer Architecture and Technology Department University of Granada Granada Spain

The concept of a small, single-layer water Cherenkov detector, with three photomultiplier tubes (PMTs), placed at its bottom in a 120° star configuration (Mercedes Water Cherenkov Detector) is presen-ted. The PMTs are placed near the lateral walls of the stations with an adjustable inclination and may be in-stalled inside or outside the water volume. To illustrate the technical viability of this concept and obtain a first-order estimation of its cost, an engineering design was elaborated. The sensitivity of these stations to low energy Extensive Air Shower (EAS) electrons, photons and muons is discussed, both in compact and sparse array configurations. It is shown that the analysis of the intensity and time patterns of the PMT signals, using machine learning techniques, enables the tagging of muons, achieving an excellent gamma/hadron discrimination for TeV showers. This concept minimises the station production and maintenance costs, allowing for a highly exible and fast installation. Mercedes Water Cherenkov Detectors (WCDs) are thus well-suited for use in high-altitude large gamma-ray observatories covering an extended energy range from the low energies, closing the gap between satellite and ground-based measurements, to very high energy regions, beyond the PeV scale. Copyright © 2022, The Authors. All rights reserved.

关键词： Tellurium compounds

Impact of Large-Scale Correlated Failures on Multilevel Virtualized Networks

学校读者我要写书评

暂无评论

Impact of Large-Scale Correlated Failures on Multilevel Virt...

IEEE Workshop on High Performance Switching and Routing

作者： Max G. Medina Mohammed J.F. Alenazi Egemen K. Çetinkaya Department of Electrical & Computer Eng. Missouri Univ. of Science and Technology Rolla MO USA Department of Computer Eng. King Saud University Riyadh Saudi Arabia Technology Architecture and Planning Verizon Basking Ridge NJ USA

ISBN: (数字)9781728148465

ISBN: (纸本)9781728148472

Communication network architectures have been evolving with increased programmability and virtualization of physical infrastructures. While this evolution enables a faster introduction of new services and efficient utilization of resources, the complexities and interdependencies among the physical and virtual infrastructures, as well as the control and management planes, hinder a complete understanding of the system at hand. It is imperative, therefore, to model and study these evolving virtualized infrastructures. We propose a multilevel virtualized network embedding model and study its performance against correlated failures. Our analysis of multiple synthetic network models identified that grid-like infrastructures capture the realistic Internet2 structure the best.

关键词： Analytical models Topology Measurement Resilience Network architecture Cloud computing

Non-convex potential games for finding global solutions to sensor network localization

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Xu, Gehui Chen, Guanpu Hong, Yiguang Fidan, Baris Parisini, Thomas Johansson, Karl H. Key Laboratory of Systems and Control Academy of Mathematics and Systems Science Beijing China Division of Decision and Control Systems School of Electrical Engineering and Computer Science KTH Royal Institute of Technology Stockholm100 44 Sweden Department of Control Science and Engineering Tongji University Shanghai201804 China Shanghai Research Institute for Intelligent Autonomous Systems Shanghai201210 China Department of Mechanical and Mechatronics Engineering University of Waterloo WaterlooONN2L 3G1 Canada Department of Electrical and Electronic Engineering Imperial College London LondonSW7 2AZ United Kingdom Department of Engineering and Architecture University of Trieste Trieste34127 Italy

Sensor network localization (SNL) problems require determining the physical coordinates of all sensors in a network. This process relies on the global coordinates of anchors and the available measurements between non-anchor and anchor nodes. Attributed to the intrinsic non-convexity, obtaining a globally optimal solution to SNL is challenging, as well as implementing corresponding algorithms. In this paper, we formulate a non-convex multi-player potential game for a generic SNL problem to investigate the identification condition of the global Nash equilibrium (NE) therein, where the global NE represents the global solution of SNL. We employ canonical duality theory to transform the non-convex game into a complementary dual problem. Then we develop a conjugation-based algorithm to compute the stationary points of the complementary dual problem. On this basis, we show an identification condition of the global NE: the stationary point of the proposed algorithm satisfies a duality relation. Finally, simulation results are provided to validate the effectiveness of the theoretical results. Copyright © 2023, The Authors. All rights reserved.

关键词： Sensor networks

Data-Driven Reachability Analysis for Piecewise Affine System

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Xie, Peng Betz, Johannes Raimondo, Davide M. Alanwar, Amr TUM School of Computation Information and Technology Department of Computer Engineering Technical University of Munich Heilbronn74076 Germany TUM School of Engineering and Design Department of Autonomous Vehicles Systems Technical University of Munich Garching85748 Germany School of Automatic Control Department of Engineering and Architecture University of Trieste Via Valerio 10 Trieste34127 Italy

Hybrid systems play a crucial role in modeling real-world applications where discrete and continuous dynamics interact, including autonomous vehicles, power systems, and traffic networks. Safety verification for these systems requires determining whether system states can enter unsafe regions under given initial conditions and uncertainties—a question directly addressed by reachability analysis. However, hybrid systems present unique difficulties because their state space is divided into multiple regions with distinct dynamic models, causing traditional data-driven methods to produce inadequate over-approximations of reachable sets at region boundaries where dynamics change abruptly. This paper introduces a novel approach using hybrid zonotopes for data-driven reachability analysis of piecewise affine systems. Our method addresses the boundary transition problem by developing computational algorithms that calculate the family of set models guaranteed to contain the true system trajectories. Additionally, we extend and evaluate three methods for set-based estimation that account for input-output data with measurement noise. © 2025, CC BY.

关键词： Convergence of numerical methods

Cloud-Magnetic Resonance Imaging System: In the Era of 6G and Artificial Intelligence

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhou, Yirong Wu, Yanhuang Su, Yuhan Li, Jing Cai, Jianyun You, Yongfu Guo, Di Qu, Xiaobo The Department of Electronic Science Biomedical Intelligent Cloud R&D Center Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance National Institute for Data Science in Health and Medicine Institute of Artificial Intelligence Xiamen University Xiamen361104 China The Department of Electronic Science Key Laboratory of Digital Fujian on IoT Communication Architecture and Safety Technology Xiamen University Xiamen361104 China Shanghai Electric Group CO. LTD Shanghai200001 China China Telecom Group Quanzhou362000 China China Mobile Group Xiamen361005 China School of Computer and Information Engineering Fujian Engineering Research Center for Medical Data Mining and Application Xiamen University of Technology Xiamen361024 China

Magnetic Resonance Imaging (MRI) plays an important role in medical diagnosis, generating petabytes of image data annually in large hospitals. This voluminous data stream requires a significant amount of network bandwidth and extensive storage infrastructure. Additionally, local data processing demands substantial manpower and hardware investments. Data isolation across different healthcare institutions hinders cross-institutional collaboration in clinics and research. In this work, we anticipate an innovative MRI system and its four generations that integrate emerging distributed cloud computing, 6G bandwidth, edge computing, federated learning, and blockchain technology. This system is called Cloud-MRI, aiming at solving the problems of MRI data storage security, transmission speed, AI algorithm maintenance, hardware upgrading, and collaborative work. The workflow commences with the transformation of k-space raw data into the standardized Imaging Society for Magnetic Resonance in Medicine Raw Data (ISMRMRD) format. Then, the data are uploaded to the cloud or edge nodes for fast image reconstruction, neural network training, and automatic analysis. Then, the outcomes are seamlessly transmitted to clinics or research institutes for diagnosis and other services. The Cloud-MRI system will save the raw imaging data, reduce the risk of data loss, facilitate inter-institutional medical collaboration, and finally improve diagnostic accuracy and work efficiency. Copyright © 2023, The Authors. All rights reserved.

关键词： Magnetic resonance imaging

Blending effects of boron carbide and lanthanum(III) oxide reinforcements on the structural, physical, mechanical, and radiation shielding properties of 316L-stainless steel composites for nuclear applications

学校读者我要写书评

暂无评论

Annals of Nuclear Energy 2025年 221卷

作者： ALMisned, Ghada Yayla, Nihal Albayrak, M.Gökhan Güler, Ömer Baykal, Duygu Sen Alkarrani, Hessa Tekin, H.O. Department of Physics College of Science Princess Nourah Bint Abdulrahman University P.O. Box 84428 Riyadh11671 Saudi Arabia Munzur University Department of Strategic Raw Materials and Advanced Technology Applications Tunceli Turkey Fırat University Metallurgical and Materials Engineering Department Engineering Faculty Elazığ Turkey Rare Earth Elements Application and Research Center Munzur University Tunceli62000 Turkey Istanbul Nisantasi University Faculty of Engineering and Architecture Mechatronics Engineering Istanbul34398 Turkey Research Institute for Medical and Health Sciences University of Sharjah Sharjah Sharjah27272 United Arab Emirates Medical Diagnostic Imaging Department College of Health Sciences University of Sharjah Sharjah27272 United Arab Emirates Istinye University Faculty of Engineering and Natural Sciences Computer Engineering Department Istanbul34396 Turkey

Four composite samples were fabricated, maintaining 10 % B4C and varying La2O3 content from 0 % to 20 %. X-ray Diffraction (XRD), Scanning Electron Microscopy (SEM), and Energy Dispersive X-ray Spectroscopy (EDX) analyses confirmed the successful incorporation of reinforcements, preserving the austenitic structure of 316L-SS. Elastic modulus calculations revealed a gradual reduction from 225.0 GPa to 215.0 GPa with increasing La2O3 due to clustering and reduced homogeneity. Gamma-ray shielding properties demonstrated significant improvements, particularly at low and intermediate photon energies, with Sample 4 with 20 % La2O3 addition, achieving a 13.6 % higher linear attenuation coefficients (LAC) at 0.015 MeV. Neutron shielding highlighted Sample 4 as one of the most effective materials, outperforming conventional alloys. This study demonstrates the potential of 316L Stainless-Steel (316L-SS) + B4C + La2O3 composites as multifunctional materials for radiation-intensive applications, offering superior attenuation and structural stability. It can be concluded that optimization of reinforcement distribution would play an important role in balancing mechanical and shielding performance in next-generation nuclear materials. © 2025 Elsevier Ltd

关键词： Reactor shielding

iLGaCo: Incremental learning of gait covariate factors

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Mu, Zihao Castro, Francisco M. Marín-Jiménez, Manuel J. Guil, Nicolás Li, Yan-Ran Yu, Shiqi College of Computer Science and Software Engineering Shenzhen University China Department of Computer Architecture University of Málaga Spain Department of Computing and Numerical Analysis University of Córdoba Spain China Department of Computer Science and Engineering Southern University of Science and Technology China

Gait is a popular biometric pattern used for identifying people based on their way of walking. Traditionally, gait recognition approaches based on deep learning are trained using the whole training dataset. In fact, if new data (classes, view-points, walking conditions, etc.) need to be included, it is necessary to re-train again the model with old and new data samples. In this paper, we propose iLGaCo, the first incremental learning approach of covariate factors for gait recognition, where the deep model can be updated with new information without re-training it from scratch by using the whole dataset. Instead, our approach performs a shorter training process with the new data and a small subset of previous samples. This way, our model learns new information while retaining previous knowledge. We evaluate iLGaCo on CASIA-B dataset in two incremental ways: adding new view-points and adding new walking conditions. In both cases, our results are close to the classical ‘training-from-scratch’ approach, obtaining a marginal drop in accuracy ranging from 0.2% to 1.2%, what shows the efficacy of our approach. In addition, the comparison of iLGaCo with other incremental learning methods, such as LwF and iCarl, shows a significant improvement in accuracy, between 6% and 15% depending on the experiment. Copyright © 2020, The Authors. All rights reserved.

关键词： Gait analysis