检索结果-内蒙古大学图书馆

5th International Conference on Information science, Parallel and Distributed Systems, ISPDS 2024

作者： Zhou, Yuhang Shen, Wenfeng Shanghai University School of Computer Engineering and Science Shanghai China Xi'an Innovation College of Yan'an University China Shanghai Polytechnic University School of Computer and Information Engineering Shanghai China

ISBN: (纸本)9798350352719

Federated Learning (FL) enables the collaborative training of a global model by distributed edge devices, without necessitating the sharing of their local datasets. Yet, FL incurs considerable communication overhead, primarily due to the server and clients' recurrent exchange of model updates, particularly from uploading gradients from the clients. To address this problem, most existing research utilizes quantization, which compresses model parameters from high-precision floating point numbers to fixed-point numbers. However, current quantization approaches may lead to underflow due to varying gradient distributions and cause the model to fail to converge. Moreover, these approaches are either fixed or unrelated to the training status, failing to adequately balance performance and efficiency. In this work, an innovative quantization algorithm, FedLAQ, is introduced, designed to solve the communication bottleneck without compromising model accuracy. Specifically, FedLAQ relates the strength of quantization to the training status using a TCP-inspired strategy, and quantization is performed at a layer level with finer and coarser granularity. Experiments demonstrate that FedLAQ significantly enhances client-to-server compression, surpassing the top-performing baselines (e.g., QSGD) by up to a factor of 1.73. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Modulation for Step-load Change of a Neutralpoint-clamped Hybrid Three-level Dual Active Bridge Converter 3

Modulation for Step-load Change of a Neutralpoint-clamped Hy...

引用

3rd International Conference on Power Systems and Electrical Technology, PSET 2024

作者： Ding, Ruiqi Liu, Shiyuan Tang, Chao Li, Xiaodong Sun, Chuan Hu, Song Macau University of Science and Technology Faculty of Innovation Engineering China School of Electrical Engineering and Automation Changshu Institute of Technology Changshu China

ISBN: (纸本)9798350353204

As a high-voltage and high-power bidirectional dc-dc-converter, the neutral-point-clamped hybrid three-level dual active bridge (NH3L-DAB) converter is a competitive candidate for the solid-state transformer in DC distribution grid. To meet the requirements of power adjustment, the converter is expected to response to the step load change quickly and stably through proper transient modulation of available multiple phase-shift angles. To eliminate the transient overshoot current and temporal dc bias in the transformer current, an optimized transient modulation method is proposed in the current work by inserting a variable compensation to the adjustment of each phase-shift angles. By forcing the transient dc average current to be zero, the unknown compensation can found to achieve an accurate and smooth transient process within a switching period. Finally, the correctness and effectiveness of the optimized modulation method are validated by both simulation results and the experimental test. © 2024 IEEE.

关键词： Phase shift

来源：评论

学校读者我要写书评

暂无评论

Optimizing Algorithms for Human Behavior Recognition: Single-Modality Spatiotemporal Convolutional Networks 11

Optimizing Algorithms for Human Behavior Recognition: Single...

引用

11th IEEE International Conference on Behavioural and Social Computing, BESC 2024

作者： Zhao, Rui Hao, Jutao Feng, Zhikun Wang, Xianzhi Prasad, Mukesh Huo, Huan School of Computer Science University of Technology Sydney Faculty of Engineering and IT Sydney Australia School of Electric Information Engineering Shanghai Dianji University Shanghai China School of Computer Science and Engineering Tianjin University of Technology Tianjin China

ISBN: (纸本)9798331531904

Behavior recognition constitutes a pivotal area of research within computer vision and serves as an integral technology in video surveillance and related domains, thereby possessing substantial research value. Convolutional neural networks (CNNs) have been proven to possess robust learning capabilities. However, due to the high complexity and variability of human behavior, recognizing human behaviors in videos remains challenging due to the difficulty of capturing useful features over long time series and the redundancy of multi-frame information. To address these issues and improve the utilization of continuous frame information, we have designed a cross-frame multiscale information fusion module. This module allows the backbone network to acquire multiscale temporal features from different frames, thereby reducing information redundancy across multiple frames while minimizing additional computational costs. Furthermore, to mitigate the problem of video temporal feature loss, we introduce an easily integrable attention mechanism module, the Easy-Attention Block. This module facilitates rapid capture of long-range dependencies, enabling the model to focus on the most relative parts of continuous video sequences and suppress irrelevant temporal information. Consequently, the model can extract more discriminative spatial-temporal features, thereby enhancing its efficiency in handling long sequence data. Experimental results express that this model can achieve high recognition accuracy and effectively extract the temporal features of video, suitable for real-time applications in critical infrastructure surveillance, which enhancing safety and reducing potential risks in energy power systems. © 2024 IEEE.

关键词： Spatio-temporal data

来源：评论

学校读者我要写书评

暂无评论

A Smart CIoT with Secure Healthcare Framework Using Optimized Deep Recuperator Neural Network Long Short-Term Memory

引用

IEEE Internet of Things Journal 2024年第6期11卷 10551-10562页

作者： Narmatha, C. Manimurugan, S. Karthikeyan, P. University of Tabuk Faculty of Computers and Information Technology Tabuk71491 Saudi Arabia Oakland University Department of Electrical and Computer Engineering RochesterMI112345 United States National Chung Cheng University Department of Computer Science and Information Engineering Chiayi62102 Taiwan RV University School of Computer Science and Engineering Bengaluru560059 India

The healthcare system currently relies on the facility to store and process large amounts of health data, supported by efficient management. The Internet of Things (IoT) has driven the growth of Adroit Healthcare, which has vast data processing capabilities and extensive data collection. The consumer IoT (CIoT), also known as the IoT in the context of individual use cases, is dominated by personal healthcare applications. The remarkable expansion of the CIoT can be attributed to the extensive embrace of wearable technology, facilitating everyday monitoring of vital health indicators like blood pressure, heart rate, and respiration rate. However, the CIoT raises concerns about data security, confidentiality, and customer trust. This study proposes an integrating blockchain and deep learning (DL) approach to promote responsible use of CIoT. The study develops an Intelligent IoT (IIoT) and healthcare diagnostic model, named BT-PWO-DRNN-LSTM, using blockchain technology and Prairie Wolf optimization-based deep recuperator neural network (DRNN) and long short-term memory (LSTM). The BT-PWO-DRNN-LSTM approach includes three key processes: 1) encrypted transactions;2) cryptographic hash feature;and 3) medical diagnosis. The PWO technique is used for secure communication of medical images, the neighborhood catalogue disposition (NCD) technique is used in the cryptographic hash feature operation (CHFO), and the DL approach is employed as a classification algorithm for diagnosing disorders. The use of PWO methodology for secure healthcare communication and optimal parameter fine-tuning highlights the originality of the study. The BT-PWO-DRNN-LSTM prototype showed positive results with high sensitivity, specificity, and accuracy during the diagnostic process. © 2014 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Chaotic Convolutional Long Short-Term Memory Network for Respiratory Motion Prediction 4

Chaotic Convolutional Long Short-Term Memory Network for Res...

引用

4th International Conference on Image Processing and Vision engineering, IMPROVE 2024

作者： Ghasemi, Narges Nabavi, Shahabedin Moghaddam, Mohsen Ebrahimi Shekofteh, Yasser Faculty of Computer Science and Engineering Shahid Beheshti University Tehran Iran Department of Computer Science Viterbi School of Engineering University of Southern California United States

ISBN: (纸本)9789897586934

One of the challenges of treating lung tumors in radiation therapy is the patient's respiratory movements during the treatment, which lead to tumor motion. The goal of respiratory motion prediction is to predict the movements of lung tissues and lung tumors during the breathing cycle. Predicting respiratory movements allows radiation to be directed only at the tumor, minimizing exposure to healthy tissue and reducing the risk of side effects. Using 4D CT images, we can find the next position of the lung tumor and make a 4D radiation therapy plan. As obtaining 4D CT scans is harmful to the patient due to radiation, the aim of this study is to construct a 4D CT during a respiratory cycle using only a 3D image. In this paper, a Chaotic Convolutional Long Short-Term Memory network is proposed, which utilizes chaotic features in respiratory signals to predict pulmonary movements more accurately. The innovation of this method is paying attention to chaotic features of respiratory signals, which leads to better interpretability of the presented model. The obtained results show that the proposed method has a higher learning speed and better performance compared to previous models, which generate 4D CT scans. Copyright © 2024 by SCITEPRESS - science and Technology Publications, Lda.

关键词： Radiotherapy

来源：评论

学校读者我要写书评

暂无评论

Unmanned aerial vehicles towards future Industrial Internet:Roles and opportunities

引用

Digital Communications and Networks 2024年第4期10卷 873-883页

作者： Linpei Li Chunlei Sun Jiahao Huo Yu Su Lei Sun Yao Huang Ning Wang Haijun Zhang School of Computer and Communication Engineering University of Science and Technology BeijingBeijing 100083China Shunde Innovation School University of Science and Technology BeijingFoshan 528399China China Mobile Chengdu Institute of Research and Development Chengdu 610200China School of Automation and Electrical Engineering University of Science and Technology BeijingBeijing 100083China Chengdu University of Information Technology Chengdu 610225China Zhengzhou University Zhengzhou 450001China

Unmanned Aerial Vehicles(UAVs)are gaining increasing attention in many fields,such as military,logistics,and hazardous site *** UAVs to assist communications is one of the promising applications and research *** future Industrial Internet places higher demands on communication *** easy deployment,dynamic mobility,and low cost of UAVs make them a viable tool for wireless communication in the Industrial ***,UAVs are considered as an integral part of Industry *** this article,three typical use cases of UAVs-assisted communications in Industrial Internet are first ***,the state-of-the-art technologies for drone-assisted communication in support of the Industrial Internet are *** to the current research,it can be assumed that UAV-assisted communication can support the future Industrial Internet to a certain ***,the potential research directions and open challenges in UAV-assisted communications in the upcoming future Industrial Internet are discussed.

关键词： Unmanned aerial vehicles(UAVs) UAV-assisted communications Industrial Internet

来源：评论

学校读者我要写书评

暂无评论

ReE3D: Boosting Novel View Synthesis for Monocular Images using Residual Encoders

引用

IEEE Transactions on Multimedia 2023年 1-11页

作者： Guo, KeHua Chen, Tianyu Ren, Sheng Hu, Bin Wu, Zheng Guo, Shaojun Fang, Hui School of Computer Science and Engineering Central South University Changsha China School of Computer and Electrical Engineering Hunan University of Arts and Science Changde China National Innovation of Defense Technology Academy of Military Sciences PLA China Beijing China Computer Science Department Loughborough University Loughborough U.K

In recent years, novel view synthesis from a monocular image has become a research hot-spot that attracts significant attention. Some recent work identifies latent vectors for high-quality view generation via iterative optimisation, which is a time-consuming process. In contrast, some others utilise an encoder learning a mapping function to approximately estimate optimal latent codes, which significantly reduces its processing time but sacrifices reconstruction quality. Consequently, how to balance synthesis quality and its generation efficiency still remains challenging. In this paper, we propose a residual-based encoder to incorporate with a 3D Generative Adversarial Networks (GAN), named ReE3D, for novel view synthesis. It applies an iterative prediction of latent codes to ensure much higher quality of novel view synthesis with an insignificant increase of processing time when compared to existing encoder-based 3D GAN inversion methods. Additionally, we enforce a novel geometric loss constraint on the encoder to predict view-invariant latent codes, thus effectively mitigating the trade-off between geometric and texture quality in 3D GAN inversion. Extensive experimental results demonstrate that our extended encoder-based method has achieved best trade-off performance in terms of novel view synthesis quality and its execution time. Our method has gained comparable synthesis quality with exponentially decreased processing time when compared to iterative optimisation methods, while improved synthesis performance of encoder-based methods significantly. IEEE

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Design of artificial biomimetic channels with Na^(+)permeation rate and selectivity potentially outperforming the natural sodium channel

引用

Nano Research 2024年第9期17卷 8638-8646页

作者： Zhi Zhu Yan Zhao Chao Chang Shaojian Yan Tingyu Sun Shiyu Gu Yangmei Li Dengsong Zhang Chunlei Wang Xiao Cheng Zeng School of Optical-Electrical and Computer Engineering University of Shanghai for Science and TechnologyShanghai 200093China Innovation Laboratory of Terahertz Biophysics National Innovation Institute of Defense TechnologyBeijing 100071China School of Physics Peking UniversityBeijing 100871China International Joint Laboratory of Catalytic Chemistry College of SciencesShanghai UniversityShanghai 200444China Department of Materials Science and Engineering City University of Hong KongHong Kong 999077China

Artificial ion channels that enable high-efficiency ion transport have important implications in nanofluidics and biomedical applications such as drug ***,we show a simulation-based chemical design of a biomimetic sodium channel that possesses permeation rate and selectivity potentially higher than those of the state-of-the-art natural vertebrate voltage-gated sodium ***,our theoretical findings have undergone empirical testing,aligning well with the Arrhenius law as derived from a diverse range of experimental *** high-efficiency ion transport is achieved by anchoring the carboxylate functional groups within the channel filter.A key chemical guiding principle underlying the ion channel design is that the free-energy barrier for the Na^(+)passage across the channel should be comparable to typical thermal energy at room *** the implementation of the chemical design,we found that the relatively low free-energy barrier can be attributed to the compensation effect of the carboxylate groups to the partially lost oxygen shell of the ion within the ion channel,as well as to the consonant vibration of the ions inside and outside the *** mechanistic understanding brings new insight,at the molecular level,into the high-efficiency ion transport across the designed membrane *** proof of principle achieved from the simulations will stimulate future experimental confirmation and potential applications of the high-performance artificial channels in nanofluidics and in bioinspired iontronics.

关键词： sodium channels anchoring carboxylate functional groups molecular dynamics simulations low free-energy barrier for ion passage thermal kinetic energy

来源：评论

学校读者我要写书评

暂无评论

A survey for light field super-resolution

引用

High-Confidence Computing 2024年第1期4卷 118-129页

作者： Mingyuan Zhao Hao Sheng Da Yang Sizhe Wang Ruixuan Cong Zhenglong Cui Rongshan Chen Tun Wang Shuai Wang Yang Huang Jiahao Shen State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and EngineeringBeihang UniversityBeijing 100191China Key Laboratory of Data Science and Intelligent Computing International Innovation InstituteBeihang UniversityHangzhou 311115China State Key Laboratory of Software Development Environment School of Computer Science and EngineeringBeihang UniversityBeijing 100191China

Compared to 2D imaging data,the 4D light field(LF)data retains richer scene’s structure information,which can significantly improve the computer’s perception capability,including depth estimation,semantic segmentation,and LF ***,there is a contradiction between spatial and angular resolution during the LF image acquisition *** overcome the above problem,researchers have gradually focused on the light field super-resolution(LFSR).In the traditional solutions,researchers achieved the LFSR based on various optimization frameworks,such as Bayesian and Gaussian *** learning-based methods are more popular than conventional methods because they have better performance and more robust generalization *** this paper,the present approach can mainly divided into conventional methods and deep learning-based *** discuss these two branches in light field spatial super-resolution(LFSSR),light field angular super-resolution(LFASR),and light field spatial and angular super-resolution(LFSASR),***,this paper also introduces the primary public datasets and analyzes the performance of the prevalent approaches on these ***,we discuss the potential innovations of the LFSR to propose the progress of our research field.

关键词： Light field super-resolution Convolutional neural network Transformer Sub-aperture image Epipolar-plane image

来源：评论

学校读者我要写书评

暂无评论

Decreasing the Design Complexity of Cyber-Physical Systems through Conceptual Modeling: The Robot Abstraction Method 10

Decreasing the Design Complexity of Cyber-Physical Systems t...

引用

10th IEEE World Forum on Internet of Things, WF-IoT 2024

作者： Voelz, Alexander Kiesenhofer, Daniel Karagiannis, Dimitris University of Vienna Faculty of Computer Science Doctoral School Computer Science Vienna Austria University of Vienna Faculty of Computer Science Research Group Knowledge Engineering Vienna Austria

ISBN: (纸本)9798350373011

The increasing complexity of designing, deploying, and maintaining Cyber-Physical Systems (CPS), particularly those incorporating multiple interacting robots, presents significant challenges regarding programming and system integration. Existing methods and tools that aim to decrease the complexity of setting up such systems through modeling-based approaches often focus on single-robot interaction and rely on manual data entries, thus limiting their scalability and applicability to more intricate multi-robot environments. This contribution introduces the Robot ion Method (RobAM), a conceptual modeling method developed to decrease the design complexity of CPS through additional layers of abstraction. By utilizing conceptual modeling, capabilities of humanoid robots can be leveraged for the design of multi-robot interactions. Thereby, our approach enhances the modeling of complex CPS as well as the subsequent generation of code to enable efficient deployments. Through an experimental scenario, we demonstrate how RobAM simplifies the development process of CPS, paving the way for advancements regarding the integration of AI-supported technologies, such as Large Language Models, to ultimately generate contextually relevant semantics for future system design. © 2024 IEEE.

关键词： Multipurpose robots

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：