检索结果-内蒙古大学图书馆

arXiv 2021年

作者： Baek, Hankyul Yun, Won Joon Kwak, Yunseok Jung, Soyi Ji, Mingyue Bennis, Mehdi Park, Jihong Kim, Joongheon Department of Electrical and Computer Engineering Korea University Seoul Korea Republic of School of Software Hallym University Chuncheon Korea Republic of Department of Electrical and Computer Engineering University of Utah Salt Lake CityUT United States Centre for Wireless Communications University of Oulu Oulu Finland School of Information Technology Deakin University Geelong Australia

This paper aims to integrate two synergetic technologies, federated learning (FL) and width-adjustable slimmable neural network (SNN) architectures. FL preserves data privacy by exchanging the locally trained models of mobile devices. By adopting SNNs as local models, FL can flexibly cope with the time-varying energy capacities of mobile devices. Combining FL and SNNs is however non-trivial, particularly under wireless connections with time-varying channel conditions. Furthermore, existing multi-width SNN training algorithms are sensitive to the data distributions across devices, so are ill-suited to FL. Motivated by this, we propose a communication and energy efficient SNN-based FL (named SlimFL) that jointly utilizes superposition coding (SC) for global model aggregation and superposition training (ST) for updating local models. By applying SC, SlimFL exchanges the superposition of multiple width configurations that are decoded as many as possible for a given communication throughput. Leveraging ST, SlimFL aligns the forward propagation of different width configurations, while avoiding the inter-width interference during back propagation. We formally prove the convergence of SlimFL. The result reveals that SlimFL is not only communication-efficient but also can counteract non-IID data distributions and poor channel conditions, which is also corroborated by simulations. © 2021, CC BY.

关键词： Backpropagation

来源：评论

学校读者我要写书评

暂无评论

Terahertz Communications and Sensing for 6G and Beyond: A Comprehensive View

TechRxiv

引用

TechRxiv 2023年

作者： Jiang, Wei Zhou, Qiuheng He, Jiguang Habibi, Mohammad Asif Melnyk, Sergiy El-Absi, Mohammed Han, Bin Di Renzo, Marco Schotten, Hans Dieter Luo, Fa-Long El-Bawab, Tarek S. Juntti, Markku Debbah, Mérouane Leung, Victor C.M. Kaiserslautern Germany The Technology Innovation Institute Abu Dhabi United Arab Emirates Centre for Wireless Communications University of Oulu Oulu90014 Finland The Institue of Digital Signal Processing University of Duisburg-Essen Duisburg Germany Germany Université Paris-Saclay CNRS CentraleSupélec Laboratoire des Signaux et Systèmes 3 Rue Joliot-Curie Gif-sur-Yvette91192 France The Electrical and Computer Engineering Department University of Washington SeattleWA98195 United States Nigeria The University of Oulu Finland Khalifa University of Science and Technology P O Box 127788 Abu Dhabi United Arab Emirates The College of Computer Science and Software Engineering Shenzhen University China Canada Germany

The next-generation wireless technologies, commonly referred to as the sixth generation (6G), are envisioned to support extreme communications capacity and in particular disruption in the network sensing capabilities. The terahertz (THz) band is one potential enabler for those due to the enormous unused frequency bands and the high spatial resolution enabled by both short wavelengths and bandwidths. Different from earlier surveys, this paper presents a comprehensive treatment and technology survey on THz communications and sensing in terms of the advantages, applications, propagation characterization, channel modeling, measurement campaigns, antennas, transceiver devices, beamforming, networking, the integration of communications and sensing, and experimental testbeds. Starting from the motivation and use cases, we survey the development and historical perspective of THz communications and sensing with the anticipated 6G requirements. We explore the radio propagation, channel modeling, and measurements for THz band. The transceiver requirements, architectures, technological challenges, and approaches together with means to compensate for the high propagation losses by appropriate antenna and beamforming solutions. We survey also several system technologies required by or beneficial for THz systems. The synergistic design of sensing and communications is explored with depth. Practical trials, demonstrations, and experiments are also summarized. The paper gives a holistic view of the current state of the art and highlights the issues and challenges that are open for further © 2023, CC BY-NC-SA.

关键词： Beamforming

来源：评论

学校读者我要写书评

暂无评论

Ticking clocks as dependent right adjoints denotational semantics for clocked type theory

arXiv

引用

arXiv 2020年

作者： Mannaa, Bassel Møgelberg, Rasmus Ejlers Veltri, Niccolò Wissmann, Thorsten eToroX Labs Denmark Department of Computer Science IT University of Copenhagen Copenhagen Denmark Department of Software Science Tallinn University of Technology Tallinn Estonia

Clocked Type Theory (CloTT) is a type theory for guarded recursion useful for programming with coinductive types, allowing productivity to be encoded in types, and for reasoning about advanced programming language features using an abstract form of step-indexing. CloTT has previously been shown to enjoy a number of syntactic properties including strong normalisation, canonicity and decidability of the equational theory. In this paper we present a denotational semantics for CloTT useful, e.g., for studying future extensions of CloTT with constructions such as path types. The main challenge for constructing this model is to model the notion of ticks on a clock used in CloTT for coinductive reasoning about coinductive types. We build on a category previously used to model guarded recursion with multiple clocks. In this category there is an object of clocks but no object of ticks, and so tick-assumptions in a context can not be modelled using standard tools. Instead we model ticks using dependent right adjoint functors, a generalisation of the category theoretic notion of adjunction to the setting of categories with families. Dependent right adjoints are known to model Fitch-style modal types, but in the case of CloTT, the modal operators constitute a family indexed internally in the type theory by clocks. We model this family using a dependent right adjoint on the slice category over the object of clocks. Finally, we show how to model the tick constant of CloTT using a semantic substitution. This work improves on a previous model by two of the authors which not only had a flaw but was also considerably more complicated. © 2020, CC BY.

关键词： Formal languages

来源：评论

学校读者我要写书评

暂无评论

Energy-Aware Resource Scheduling with Fault-Tolerance in Edge Computing 16th

Energy-Aware Resource Scheduling with Fault-Tolerance in Edg...

引用

16th IFIP WG 10.3 International Conference on Network and Parallel Computing, NPC 2019

作者： Xue, Yanfen Fan, Guisheng Yu, Huiqun Sun, Huaiying Department of Computer Science and Engineering East China University of Science and Technology Shanghai China Shanghai Key Laboratory of Computer Software Evaluating and Testing Shanghai China

ISBN: (纸本)9783030307080

Edge computing extends computation and storage resources to the edge of the network, which largely improve the performance problem of cloud computing incurred by the bandwidth limitation. And it still needs to address the challenges of energy and reliability. In this paper, we propose an energy-aware fault-tolerant resource scheduling algorithm to improve system reliability while minimizing the energy consumption. We allocate resources by reliability and energy-aware resource scheduling method for tasks firstly. Then, CPU temperature prediction and time between failures (TBF) prediction are used to trigger proactive fault tolerance mechanism (VM migration). The experimental results show that the reliability is greatly improved and energy consumption generated by VM migration is not very large compared to other methods. © 2019, IFIP International Federation for Information Processing.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Multilayer Fuzzy Neural Networks for Image Clustering

SSRN

引用

SSRN 2022年

作者： Wang, Yifan Ishibuchi, Hisao Er, Meng Joo Zhu, Jihua School of Software Engineering Xi’an Jiaotong University Xi’an710049 China Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen518055 China Guangdong Provincial Key Laboratory of Brain-inspired Intelligent Computation China College of Marine Electrical Engineering Dalian Maritime University Dalian116026 China Institute of Artificial Intelligence Marine Robotics China

Nowadays, labeling a huge number of images is still a very challenging task. To tackle the problem of unlabeled data, unsupervised learning has been proposed. Among many unsupervised learning algorithms, K-means is the most popular algorithm. However, in a low-dimensional space, fuzzy c-means, which is more robust and less sensitive to initialization, has several advantages over K-means clustering. On the other hand, stacked convolutional pooling structures and manifold representation play pivotal roles in image clustering. In this paper, we proposed an unsupervised multilayer fuzzy neural network for image clustering, which unifies fuzzy systems, multilayer convolutional structures and manifold representation. The main contributions of this work are: First, we extend fuzzy systems to unsupervised tasks by introducing the manifold representation. Next, we proposed the idea of using only a small number of attributes with the largest variances to compute firing strengths. This is to prevent the firing strengths from falling to zero in clustering high-dimensional data. Then, randomly generated convolutional weights are used to extract features, which is a good choice for data without labels. It is demonstrated on a wide range of image datasets that the proposed approach is superior to existing fuzzy and non-fuzzy clustering algorithms. © 2022, The Authors. All rights reserved.

关键词： Fuzzy neural networks

来源：评论

学校读者我要写书评

暂无评论

Fog-cloud task scheduling of energy consumption optimisation with deadline consideration

Fog-cloud task scheduling of energy consumption optimisation...

引用

International Journal of Internet Manufacturing and Services

作者： Xu, Jiuyun Sun, Xiaoting Zhang, Ruru Liang, Hongliang Duan, Qiang School of Computer and Communication Engineering China University of Petroleum Qingdao China Software Technology Company No. 58 Kunshan Road Science and Technology City Suzhou High-Tech Zone Jiangsu Province China Department of Informatics Beijing University of Posts and Telecommunications Beijing China Information Sciences and Technology Department Pennsylvania State University PennsylvaniaPA United States

The emerging IoT introduces many new challenges that cannot be adequately addressed by the current 'cloud-only' architectures. The cooperation of the fog and cloud is considered to be a promising architecture, which efficiently handles IoT's data processing and communications requirements. However, how to schedule tasks to better adapt to IoT real-time needs and reduce the energy in the fog-cloud system is not well addressed. In this paper, we first model the energy consumption of the fog and cloud, respectively, and formulate a task scheduling problem into a constrained optimisation problem in fog-cloud computing system. Then, an efficient deadline-energy scheduling algorithm based on ant colony optimisation (DEACO) is put forward to tackle this problem, which achieves to reduce energy consumption on the condition of satisfying the task deadline. Finally, algorithms have been simulated on the extended CloudSim simulator. The experimental results have shown that our scheduling approach reduces energy more effective. © 2020 Inderscience Enterprises Ltd.

关键词： Green computing Ant colony optimization computer architecture Constrained optimization Data handling Energy utilization Fog Internet of things Multitasking Cloud systems Cloudsim Constrained optimisation problems Optimisations Real time Reduce energy consumption Task scheduling problem Task scheduling

来源：评论

学校读者我要写书评

暂无评论

Is Bigger Data Better for Defect Prediction: Examining the Impact of Data Size on Supervised and Unsupervised Defect Prediction 16th

Is Bigger Data Better for Defect Prediction: Examining the I...

引用

16th Web Information Systems and Applications Conference, WISA 2019

作者： Liu, Xinyue Li, Yanhui State Key Laboratory for Novel Software Technology Nanjing University Nanjing China Department of Computer Science and Technology Nanjing University Nanjing210023 China

ISBN: (纸本)9783030309510

Defect prediction could help software practitioners to predict the future occurrence of bugs in the software code regions. In order to improve the accuracy of defect prediction, dozens of supervised and unsupervised methods have been put forward and achieved good results in this field. One limiting factor of defect prediction is that the data size of defect data is not big, which restricts the scope of application with defect prediction models. In this study, we try to construct bigger defect datasets by merging available datasets with the same measurement dimension and check whether bigger data will bring better defect prediction performance with supervised and unsupervised models or not. The results of our experiment reveal that larger-scale dataset doesn’t bring improvements of both supervised and unsupervised classifiers. © 2019, Springer Nature Switzerland AG.

关键词： Defects

来源：评论

学校读者我要写书评

暂无评论

Dynamic Sample Selection for Federated Learning with Heterogeneous Data in Fog Computing

Dynamic Sample Selection for Federated Learning with Heterog...

引用

IEEE International Conference on Communications (ICC)

作者： Lingshuang Cai Di Lin Jiale Zhang Shui Yu Department of Information and Software Engineering University of Electronic Science and Technology Chengdu China Department of Computer Science and Technology Nanjin University of Aeronautics and Astronautics Nanjing China Department of Engineering and Information Technology University of Technology Sydney Australia

ISBN: (数字)9781728150895

ISBN: (纸本)9781728150901

Federated learning is a state-of-the-art technology used in the fog computing, which allows distributed learning to train cross-device data while achieving efficient performance. Many current works have optimized the federated learning algorithm in homogeneous networks. However, in the actual application scenario of distributed learning, data is independently generated by each device, and this non-homologous data has different distribution characteristics. Therefore, the data used by each device for local learning is unbalanced and non-IID, and the heterogeneity of data affects the performance of federated learning and slows down the convergence. In this paper, we present a dynamic sample selection optimization algorithm, FedSS, to tackle heterogeneous data in federated learning. FedSS dynamically selects the training sample size during the gradient iteration based on the locally available data size, to settle the expensive evaluations of the local objective function with a massive amount of dataset. We theoretically analyze the convergence and present the complexity estimates of our framework when learning large data from unbalanced distribution. Our experimental results show that the use of dynamic sampling methods can effectively improve the convergence speed with heterogeneous data, and keep computational costs low while achieving the desired accuracy.

关键词： Training Performance evaluation Distributed databases Convergence Computational modeling Machine learning Edge computing

来源：评论

学校读者我要写书评

暂无评论

TEA: Temporal excitation and aggregation for action recognition

arXiv

引用

arXiv 2020年

作者： Li, Yan Ji, Bin Shi, Xintian Zhang, Jianguo Kang, Bin Wang, Limin Tencent State Key Laboratory for Novel Software Technology Nanjing University China Department of Computer Science and Engineering Southern University of Science and Technology China

Temporal modeling is key for action recognition in videos. It normally considers both short-range motions and long-range aggregations. In this paper, we propose a Temporal Excitation and Aggregation (TEA) block, including a motion excitation (ME) module and a multiple temporal aggregation (MTA) module, specifically designed to capture both short- and long-range temporal evolution. In particular, for short-range motion modeling, the ME module calculates the feature-level temporal differences from spatiotemporal features. It then utilizes the differences to excite the motion-sensitive channels of the features. The long-range temporal aggregations in previous works are typically achieved by stacking a large number of local temporal convolutions. Each convolution processes a local temporal window at a time. In contrast, the MTA module proposes to deform the local convolution to a group of sub-convolutions, forming a hierarchical residual architecture. Without introducing additional parameters, the features will be processed with a series of sub-convolutions, and each frame could complete multiple temporal aggregations with neighborhoods. The final equivalent receptive field of temporal dimension is accordingly enlarged, which is capable of modeling the long-range temporal relationship over distant frames. The two components of the TEA block are complementary in temporal modeling. Finally, our approach achieves impressive results at low FLOPs on several action recognition benchmarks, such as Kinetics, Something-Something, HMDB51, and UCF101, which confirms its effectiveness and efficiency. Copyright © 2020, The Authors. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

ImplantFormer: Vision Transformer based Implant Position Regression Using Dental CBCT Data

arXiv

引用

arXiv 2022年

作者： Yang, Xinquan Li, Xuguang Li, Xuechen Wu, Peixi Shen, Linlin Deng, Yongqiang College of Computer Science and Software Engineering Shenzhen University Shenzhen China AI Research Center for Medical Image Analysis and Diagnosis Shenzhen University Shenzhen China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China Department of Stomatology Shenzhen University General Hospital Shenzhen China The School of Dentistry Shenzhen University Shenzhen China

Implant prosthesis is the most appropriate treatment for dentition defect or dentition loss, which usually involves a surgical guide design process to decide the implant position. However, such design heavily relies on the subjective experiences of dentists. In this paper, a transformer-based Implant Position Regression Network, ImplantFormer, is proposed to automatically predict the implant position based on the oral CBCT data. We creatively propose to predict the implant position using the 2D axial view of the tooth crown area and fit a centerline of the implant to obtain the actual implant position at the tooth root. Convolutional stem and decoder are designed to coarsely extract image features before the operation of patch embedding and integrate multi-level feature maps for robust prediction, respectively. As both long-range relationship and local features are involved, our approach can better represent global information and achieves better location performance. Extensive experiments on a dental implant dataset through five-fold cross-validation demonstrated that the proposed ImplantFormer achieves superior performance than existing methods. Copyright © 2022, The Authors. All rights reserved.

关键词： Forecasting

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：