检索结果-内蒙古大学图书馆

作者： Amrouch, Hussam Zervakis, Georgios Salamin, Sami Kattan, Hammam Anagnostopoulos, Iraklis Henkel, Jorg in the Computer Science Electrical Engineering Faculty University of Stuttgart Stuttgart Germany Chair for Embedded Systems Karlsruhe Institute of Technology Karlsruhe Germany Department of Electrical and Computer Engineering Southern Illinois University CarbondaleIL United States

Neural processing units (NPUs) are becoming an integral part in all modern computing systems due to their substantial role in accelerating neural networks (NNs). The significant improvements in cost-energy-performance stem from the massive array of multiply accumulate (MAC) units that remarkably boosts the throughput of NN inference. In this work, we are the first to investigate the thermal challenges that NPUs bring, revealing how MAC arrays, which form the heart of any NPU, impose serious thermal bottlenecks to on-chip systems due to their excessive power densities. For the first time, we explore: 1) the effectiveness of precision scaling and frequency scaling (FS) in temperature reductions and 2) how advanced on-chip cooling using superlattice thin-film thermoelectric (TE) open doors for new tradeoffs between temperature, throughput, cooling cost, and inference accuracy in NPU chips. Our work unveils that hybrid thermal management, which composes different means to reduce the NPU temperature, is a key. To achieve that, we propose and implement PFS-TE technique that couples precision and FS together with superlattice TE cooling for effective NPU thermal management. Using commercial signoff tools, we obtain accurate power and timing analysis of MAC arrays after a full-chip design is performed based on 14-nm Intel FinFET technology. Then, multiphysics simulations using finite-element methods are carried out for accurate heat simulations in the presence and absence of on-chip cooling. Afterward, comprehensive design-space exploration is presented to demonstrate the Pareto frontier and the existing tradeoffs between temperature reductions, power overheads due to cooling, throughput, and inference accuracy. Using a wide range of NNs trained for image classification, experimental results demonstrate that our novel NPU thermal management increases the inference efficiency (TOPS/Joule) by $1.33\times $ , $1.87\times $ , and $2\times $ under different temperature constrain

关键词： Temperature control

来源：评论

学校读者我要写书评

暂无评论

Unveiling Hidden Energy Anomalies: Harnessing Deep Learning to Optimize Energy Management in Sports Facilities

arXiv

引用

arXiv 2024年

作者： Fadli, Fodil Himeur, Yassine Elnour, Mariam Amira, Abbes Department of Architecture & Urban Planning Qatar University Doha Qatar College of Engineering and Information Technology University of Dubai Dubai United Arab Emirates Department of Electrical Engineering Qatar University Doha Qatar Department of Computer Science University of Sharjah United Arab Emirates Institute of Artificial Intelligence De Montfort University Leicester United Kingdom

Anomaly detection in sport facilities has gained significant attention due to its potential to promote energy saving and optimizing operational efficiency. In this research article, we investigate the role of machine learning, particularly deep learning, in anomaly detection for sport facilities. We explore the challenges and perspectives of utilizing deep learning methods for this task, aiming to address the drawbacks and limitations of conventional approaches. Our proposed approach involves feature extraction from the data collected in sport facilities. We present a problem formulation using Deep Feedforward Neural Networks (DFNN) and introduce threshold estimation techniques to identify anomalies effectively. Furthermore, we propose methods to reduce false alarms, ensuring the reliability and accuracy of anomaly detection. To evaluate the effectiveness of our approach, we conduct experiments on aquatic center dataset at Qatar University. The results demonstrate the superiority of our deep learning-based method over conventional techniques, highlighting its potential in real-world applications. Typically, 94.33% accuracy and 92.92% F1-score have been achieved using the proposed scheme. © 2024, CC BY.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Unsupervised path representation learning with curriculum negative sampling

arXiv

引用

arXiv 2021年

作者： Yang, Sean Bin Guo, Chenjuan Hu, Jilin Tang, Jian Yang, Bin Department of Computer Science Aalborg University Denmark Mila-Quebec AI Institute HEC CIFAR AI Research Chair Montreal Canada

Path representations are critical in a variety of transportation applications, such as estimating path ranking in path recommendation systems and estimating path travel time in navigation systems. Existing studies often learn task-specific path representations in a supervised manner, which require a large amount of labeled training data and generalize poorly to other tasks. We propose an unsupervised learning framework Path InfoMax (PIM) to learn generic path representations that work for different downstream tasks. We first propose a curriculum negative sampling method, for each input path, to generate a small amount of negative paths, by following the principles of curriculum learning. Next, PIM employs mutual information maximization to learn path representations from both a global and a local view. In the global view, PIM distinguishes the representations of the input paths from those of the negative paths. In the local view, PIM distinguishes the input path representations from the representations of the nodes that appear only in the negative paths. This enables the learned path representations encode both global and local information at different scales. Extensive experiments on two downstream tasks, ranking score estimation and travel time estimation, using two road network datasets suggest that PIM significantly outperforms other unsupervised methods and is also able to be used as a pre-training method to enhance supervised path representation learning. Copyright © 2021, The Authors. All rights reserved.

关键词： Curricula

来源：评论

学校读者我要写书评

暂无评论

Edge Computing-Based Joint Client Selection and Networking Scheme for Federated Learning in Vehicular IoT

引用

China Communications 2021年第6期18卷 39-52页

作者： Wugedele Bao Celimuge Wu Siri Guleng Jiefang Zhang Kok-Lim Alvin Yau Yusheng Ji School of computer science and information engineering Hohhot Minzu CollegeHohhot 010051China Graduate School of Informatics and Engineering The University of Electro-Communications1-5-1ChofugaokaChofu-shiTokyo182-8585 Japan Institute of Intelligent Media Technology Communication University of ZhejiangZhejiang 310018China School of Science and Technology Sunway University5Jalan UniversitiBandar Sunway47500 Petaling JayaSelangorMalaysia Information Systems Architecture Research Division National Institute of Informatics2-1-2HitotsubashiChiyoda-kuTokyo 101-8430 Japan

In order to support advanced vehicular Internet-of-Things(IoT)applications,information exchanges among different vehicles are required to find efficient solutions for catering to different application requirements in complex and dynamic vehicular *** learning(FL),which is a type of distributed learning technology,has been attracting great interest in recent years as it performs knowledge exchange among different network entities without a violation of user ***,client selection and networking scheme for enabling FL in dynamic vehicular environments,which determines the communication delay between FL clients and the central server that aggregates the models received from the clients,is still *** this paper,we propose an edge computing-based joint client selection and networking scheme for vehicular *** proposed scheme assigns some vehicles as edge vehicles by employing a distributed approach,and uses the edge vehicles as FL clients to conduct the training of local models,which learns optimal behaviors based on the interaction with *** clients also work as forwarder nodes in information sharing among network *** client selection takes into account the vehicle velocity,vehicle distribution,and the wireless link connectivity between vehicles using a fuzzy logic algorithm,resulting in an efficient learning and networking *** use computer simulations to evaluate the proposed scheme in terms of the communication overhead and the information covered in learning.

关键词： vehicular IoT federated learning client selection networking scheme

来源：评论

学校读者我要写书评

暂无评论

Hybrid Global Flexible Power Point Tracking Algorithm for Photovoltaic Systems

Hybrid Global Flexible Power Point Tracking Algorithm for Ph...

引用

European Conference on Power Electronics and Applications

作者： Anusha Kumaresan Hossein Dehghani Tafti Glen G. Farivar Josep Pou Kai Strunz Dusit Niyato Energy Research Institute at NTU Interdisciplinary Graduate Programme Nanyang Technological University Singapore Engineering Institute of Technology (EIT) Australia Department of Electrical and Electronic Engineering The University of Melbourne Victoria Australia School of Electrical and Electronic Engineering Nanyang Technological University Singapore Chair for Sustainable Electric Networks and Sources of Energy Technische Universität Berlin Germany School of Computer Science and Engineering Nanyang Technological University Singapore

ISBN: (数字)9798350364446

ISBN: (纸本)9798350364453

Frequency support in grid-connected photovoltaic (PV) systems is achieved by employing flexible power point tracking (FPPT) algorithms. Subsequently, according to the grid frequency requirements, the PV power is curtailed to a power reference, unlike the traditional method of operating the PV at the maximum power point. In practical cases, the PV arrays are prone to partial shading, and hence, there is a requirement to formulate the FPPT algorithms that operates effectively during such cases. Moreover, fast transient response is a necessity, which implies that the algorithms proposed in this paper is desirable. This paper proposes a hybrid global FPPT (GFPPT) algorithm based on secant and Newton’s method, to shift the PV operating point to the required power reference. The Newton’s method offers quadratic convergence rate and the secant method guarantees convergence. Hence, a hybrid algorithm consisting of these two methods will provide both fast and guaranteed convergence. Validation through simulation is presented in this paper and the proposed method is compared with the recently presented perturb and observe (P&O) GFPPT algorithm in the literature.

关键词： Photovoltaic systems Transient response Europe Voltage Hybrid power systems Convergence

来源：评论

学校读者我要写书评

暂无评论

DAMS:A Distributed Analytics Metadata Schema

引用

Data Intelligence 2021年第4期3卷 528-547页

作者： Sascha Welten Laurenz Neumann Yeliz Ucer Yediel Luiz Olavo Bonino da Silva Santos Stefan Decker Oya Beyan Chair Informatik 5 RWTH Aachen University52056 AachenGermany Fraunhofer Institute for Applied Information Techniques(FIT) 53757 Sankt AugustinGermany Faculty of Electrical Engineering Mathematics and Computer ScienceUniversity of Twente7500AE EnschedeThe Netherlands Department of Human Genetics Leiden University Medical CentreLeiden 2333 ZAThe Netherlands Institute of Medical Information Faculty of Medicine&University Hospital CologneUniversity of Cologne50674 CologneGermany

In recent years,implementations enabling Distributed Analytics(DA)have gained considerable attention due to their ability to perform complex analysis tasks on decentralised data by bringing the analysis to the *** concepts propose privacy-enhancing alternatives to data centralisation approaches,which have restricted applicability in case of sensitive data due to ethical,legal or social ***,the immanent problem of DA-enabling architectures is the black-box-alike behaviour of the highly distributed components originating from the lack of semantically enriched descriptions,particularly the absence of basic metadata for data sets or analysis *** approach the mentioned problems,we propose a metadata schema for DA infrastructures,which provides a vocabulary to enrich the involved entities with descriptive *** initially perform a requirement analysis with domain experts to reveal necessary metadata items,which represents the foundation of our ***,we transform the obtained domain expert knowledge into user stories and derive the most significant semantic *** the final step,we enable machine-readability via RDF(S)and SHACL *** deploy our schema in a proof-of-concept monitoring dashboard to validate its contribution to the transparency of DA ***,we evaluate the schema’s compliance with the FAIR *** evaluation shows that the schema succeeds in increasing transparency while being compliant with most of the FAIR *** a common metadata model is critical for enhancing the compatibility between multiple DA infrastructures,our work lowers data access and analysis *** represents an initial and infrastructure-independent foundation for the FAIRification of DA and the underlying scientific data management.

关键词： Distributed analytics Federated analytics Personal Health Train Metadata schema RDF(S) SHACL FAIR

来源：评论

学校读者我要写书评

暂无评论

Modern Manufacturing for Alloy Wheel Defect Detection using Image Processing and Application

Modern Manufacturing for Alloy Wheel Defect Detection using ...

引用

IEEE International Conference on computer Communication and the Internet (ICCCI)

作者： Tuaniai Archevapanich Woranidtha Krungseanmuang Vasutorn Chaowalittawin Posathip Sathaporn Punyisa Chaowalittawin Boonchana Purahong Department of Electronics and Communication Engineering Faculty of Engineering and Architecture Rajamagala University of Technology Suvarnabhumi Nonthaburi Thailand Department of Robotics and Computational Intelligent Systems School of Engineering King Mongkut's Institute of Technology Ladkrabang Bangkok Thailand Department of Indusstrail Physics School of Science Bangkok Thailand Department of Computer Engineering School of Engineering King Mongkut's Institute of Technology Ladkrabang Bangkok Thailand

ISBN: (数字)9798350351767

ISBN: (纸本)9798350351774

This paper presents an innovative approach to identifying defects in alloy wheel production by integrating image processing techniques with a mobile application platform. The system receives X-ray alloy images from the factory via mobile phone, processes them using image processing techniques to enhance clarity and readiness for defect detection, and then transmits the processed images to a Django framework via a uniform resource locator (URL). Subsequently, the system detects defects in the images, encodes them in Base64 format, and sends them to the mobile application through an API (Application Program Interface) for display on the user interface. This well-designed system architecture offers manufacturers a comprehensive solution to ensure product quality, reduce costs, and enhance customer satisfaction.

关键词： Uniform resource locators Image processing Metals Wheels Systems architecture User interfaces Mobile applications

来源：评论

学校读者我要写书评

暂无评论

Measuring Water Ponding Time, Location and Connectivity on Soil Surfaces Using Time-Lapse Images and Deep Learning

SSRN

引用

SSRN 2024年

作者： Pereira Zamboni, Pedro Alberto Blümlein, Mikesch Lenz, Jonas Gonçalves, Wesley Nunes Junior, Jose Marcato Wöhling, Thomas Eltner, Anette Institute of Photogrammetry and Remote Sensing Dresden University of Technology Dresden01069 Germany IPROconsult GmbH Dresden01069 Germany Faculty of Engineering Geography and Architecture Federal University of Mato Grosso do Sul Brazil Faculty of Computer Science Federal University of Mato Grosso do Sul Campo Grande79070-900 Brazil Institute of Hydrology and Meteorology Dresden University of Technology Dresden01069 Germany

Rainfall simulations are an established method to gain knowledge on small-scale hydrological processes like infiltration, ponding and the formation of surface runoff. Due to limitations in measuring methods, these processes must usually be understood to happen homogeneously within the bounded plot area while it is well known that they actually vary on a subplot scale. Within this study we took high resolution time-lapse images of several plots to observe and quantify the subplot processes of ponding and the formation of connectivity and surface *** investigated the potential of deep learning in the segmentation of water ponding areas in time-lapse images during rainfall simulations and to estimate the ponding time. We trained three different Convolutional Neural Networks (CNNs), considering classification uncertainty and imbalance of the ground-truth data (water pixels) as well as ensemble modeling and spatial correlation between samples. Our findings suggest that addressing ground-truth annotation uncertainty and imbalance was more important in our study than the choice of the CNNs itself, and ensemble models increase the model performance leading to more robust predictions. Overall, our results suggest that CNNs have a great potential to segment ponding areas, and thus it is possible to observe their spatio-temporal *** comparing the evolution of water ponding areas to runoff, different behaviors across the plots were observable, which could be related to differences in initial soil moisture and infiltration behaviors. Further, our image-based deep learning approach allows for direct measurement of the ponding time, and can be considered a first step to spatially and temporally resolved mapping of infiltration rates. © 2024, The Authors. All rights reserved.

关键词： Soil moisture

来源：评论

学校读者我要写书评

暂无评论

A differential public PUF design for lightweight authentication 16

A differential public PUF design for lightweight authenticat...

引用

2021 International Conference on Synthesis, Modeling, Analysis and Simulation Methods and Applications to Circuit Design, SMACD 2021 and 16th Conference on PhD Research in Microelectronics and Electronics, PRIME 2021

作者： Duan, Shengyu Sai, Gaole School of Computer Engineering and Science Shanghai University Shanghai200444 China State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences China Guangdong Provincial Key Lab of Robotics and Intelligent System Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences China CAS Key Laboratory of Human-Machine Intelligence-Synergy Systems Shenzhen Institutes of Advanced Technology China

ISBN: (纸本)9783800755899

Physical Unclonable Functions (PUFs) have emerged as a promising primitive to provide a hardware keyless security mechanism for integrated circuit applications. Public PUFs (PPUFs) address the crucial PUF vulnerability of requiring databases to store the secrete reference information. This paper investigates one of the PPUF families, differential PPUF (dPPUF), which exploits complexly interacted logic to create time gap between actual execution and simulation, and to prevent simulation attacks. Conventional dPPUFs have high costs. We observe the repressers have large impact on simulation time, but take a great proportion in current dPPUF structure. We thereby present a represser network with specially designed interconnection and shared logic, to maintain a high circuit complexity and to reduce the area cost. We show the proposed dPPUF has similar resilience over simulation attacks as the conventional design. Area cost can be reduced by up to 44%, and thus the proposed design can be used for lightweight authentication. © VDE VERLAG GMBH.

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

Communication Lower Bounds of Convolutions in CNNs 20

Communication Lower Bounds of Convolutions in CNNs

引用

32nd ACM Symposium on Parallelism in Algorithms and architectures, SPAA 2020

作者： Zhang, Xiaoyang Xiao, Junmin Tan, Guangming State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Science Beijing China

ISBN: (纸本)9781450369350

Convolution is the most time-consuming part in the computation of convolutional neural networks (CNNs). Due to the complex data dependency and the increase in the amount of model samples, the convolution suffers from high overhead on data movement. This work provides comprehensive analysis and methodologies to minimize the communication for the convolutions in CNNs. With an in-depth analysis on the I/O complexity theory under the red-blue pebble game model, we develop a general communication lower bound theory for a composite algorithm which consists of several different sub-computations. Based on the proposed theory, we establish the data movement lower bound results for three main convolution algorithms in CNNs, which are the direct convolution, the image2col method and Winograd algorithm. Furthermore, derived from I/O lower bound results, we design the near communication-optimal strategies respectively for the three main convolution algorithms by fully exploiting the data reuse. The deep analysis demonstrates that our designs are able to nearly reach the minimum communication in a two-level memory hierarchy. © 2020 Owner/Author.

关键词： Convolution

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：