检索结果-内蒙古大学图书馆

Hierarchical Continual Learning for Domain-Knowledge Retention in Healthcare Federated Learning

IEEE Transactions on Consumer Electronics 2025年

作者： Iqbal, Saeed Zhong, Xiaopin Khan, Muhammad Attique Wu, Zongze AlHammadi, Dina Abdulaziz Liu, Weixiang Choudhry, Imran Arshad Shenzhen University College of Mechatronics and Control Engineering Shenzhen518060 China Prince Mohammad Bin Fahd University Department of AI Alkhobar Saudi Arabia Princess Nourah bint Abdulrahman University College of Computer and Information Sciences Department of Information Systems Riyadh11671 Saudi Arabia University of Central Punjab Faculty of Information Technology and Computer Science Department of Software Engineering Lahore Pakistan

Internet of Medical Things (IoMT) applications encounter issues with data protection, continual adaptation, and domain-specific knowledge retention, especially in consumer-centric IoMT scenarios. We overcome these obstacles and facilitate effective knowledge retention and task adaptation in IoMT applications. This study attempts to create a unique privacy-preserving federated learning framework that combines a hierarchical learning structure with Continual Learning (CL). Despite the advancements in Federated Learning (FL), current models have trouble integrating changing datasets in real-time while protecting privacy, as well as catastrophic forgetting, which occurs when previously learned knowledge is lost when adjusting to new tasks. We present a hierarchical learning framework that makes use of three levels of models -Junior Model (JM), Consultant Model (CM), and Senior Consultant Model (SCM) -to overcome these drawbacks. Each level of the model aids in archived retention and domain-knowledge adaptation. To guarantee that the model maintains valuable information over time and adapts to new tasks with ease, our method blends domain adaptation strategies with ongoing learning approaches like knowledge distillation and elastic weight consolidation (EWC). We compare the suggested methodology with current state-of-the-art (SOTA) models on healthcare datasets for tasks like illness diagnosis and medical image categorization. According to our findings, the hierarchical continual learning model performs better than SOTA techniques in terms of accuracy, task adaptability, and privacy protection. In the healthcare industry, our study sets a new standard for privacy-preserving, continuously adaptable federated learning systems, allowing for real-time, scalable IoMT applications that can adapt dynamically to a variety of changing datasets. © 1975-2011 IEEE.

关键词： Patient treatment

来源：评论

学校读者我要写书评

暂无评论

Geo2ComMap: Deep Learning-Based MIMO Throughput Prediction Using Geographic Data

引用

IEEE Wireless Communications Letters 2025年

作者： Lin, Fan-Hao Huang, Tzu-Hao Wen, Chao-Kai Duong, Trung Q. National Sun Yat-sen University Institute of Communications Engineering Kaohsiung80424 Taiwan National Yang Ming Chiao Tung University Institute of Electrical Control Engineering Hsinchu300 Taiwan Memorial University Faculty of Engineering and Applied Science St. John'sNLA1C 5S7 Canada Queen's University Belfast School of Electronics Electrical Engineering and Computer Science Belfast United Kingdom

Accurate communication performance prediction is crucial for wireless applications such as network deployment and resource management. Unlike conventional systems with a single transceiver antenna, throughput (Tput) estimation in antenna array-based multiple-output multiple-input (MIMO) systems is computationally intensive, i.e., requiring analysis of channel matrices, rank conditions, and spatial channel quality. These calculations impose significant computational and time burdens. This paper introduces Geo2ComMap, a deep learning-based framework that leverages geographic databases to efficiently estimate multiple communication metrics across an entire area in MIMO systems using only sparse measurements. To mitigate extreme prediction errors, we propose a sparse sampling strategy. Extensive evaluations demonstrate that Geo2ComMap accurately predicts full-area communication metrics, achieving a median absolute error of 27.35 Mbps for Tput values ranging from 0 to 1900 Mbps. © 2012 IEEE.

关键词： Antenna arrays

来源：评论

学校读者我要写书评

暂无评论

Interleaved Block-based Learned Image Compression with Feature Enhancement and Quantization Error Compensation

arXiv

引用

arXiv 2025年

作者： Jiang, Shiqi Yuan, Hui Li, Shuai Hamzaoui, Raouf Wang, Xu Huo, Junyan School of software Shandong University Shandong Jinan China School of Control Science and Engineering Shandong University Shandong Jinan China School of Engineering and Sustainable Development De Montfort University LeicesterLE1 9BH United Kingdom College of Computer Science and Software Engineering Shenzhen University Shenzhen China School of Telecommunications Engineering Xidian University Xi’an China

In recent years, learned image compression (LIC) methods have achieved significant performance improvements. However, obtaining a more compact latent representation and reducing the impact of quantization errors remain key challenges in the field of LIC. To address these challenges, we propose a feature extraction module, a feature refinement module, and a feature enhancement module. Our feature extraction module shuffles the pixels in the image, splits the resulting image into sub-images, and extracts coarse features from the sub-images. Our feature refinement module stacks the coarse features and uses an attention refinement block composed of concatenated three-dimensional convolution residual blocks to learn more compact latent features by exploiting correlations across channels, within sub-images (intra-sub-image correlations), and across sub-images (inter-sub-image correlations). Our feature enhancement module reduces information loss in the decoded features following quantization. We also propose a quantization error compensation module that mitigates the quantization mismatch between training and testing. Our four modules can be readily integrated into state-of-the-art LIC methods. Experiments show that combining our modules with Tiny-LIC outperforms existing LIC methods and image compression standards in terms of peak signal-to-noise ratio (PSNR) and multi-scale structural similarity (MS-SSIM) on the Kodak dataset and the CLIC dataset. Copyright © 2025, The Authors. All rights reserved.

关键词： Image compression

来源：评论

学校读者我要写书评

暂无评论

LSU-NET: LIGHTWEIGHT AUTOMATIC ORGANS SEGMENTATION NETWORK FOR MEDICAL IMAGES

arXiv

引用

arXiv 2025年

作者： Ding, Yujie Teng, Shenghua Li, Zuoyong Chen, Xiao College of Electronic and Information Engineering Shandong University of Science and Technology Qingdao266590 China Fujian Provincial Key Laboratory of Information Processing and Intelligent Control School of Computer and Big Data Minjiang University Fuzhou350121 China

UNet and its variants have widespread applications in medical image segmentation. However, the substantial number of parameters and computational complexity of these models make them less suitable for use in clinical settings with limited computational resources. To address this limitation, we propose a novel Lightweight Shift U-Net (LSU-Net). We integrate the Light Conv Block and the Tokenized Shift Block in a lightweight manner, combining them with a dynamic weight multi-loss design for efficient dynamic weight allocation. The Light Conv Block effectively captures features with a low parameter count by combining standard convolutions with depthwise separable convolutions. The Tokenized Shift Block optimizes feature representation by shifting and capturing deep features through a combination of the Spatial Shift Block and depthwise separable convolutions. Dynamic adjustment of the loss weights at each layer approaches the optimal solution and enhances training stability. We validated LSU-Net on the UWMGI and MSD Colon datasets, and experimental results demonstrate that LSU-Net outperforms most state-of-the-art segmentation architectures. Copyright © 2025, 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

Distributed Prescribed-Time Observer for Nonlinear Systems

arXiv

引用

arXiv 2025年

作者： de Heij, Vincent Niazi, M.B. Umar Johansson, Karl H. Ahmed, Saeed Faculty of Science and Engineering University of Groningen Groningen9747 AG Netherlands Division of Decision and Control Systems Digital Futures KTH Royal Institute of Technology StockholmSE-100 44 Sweden Laboratory for Information and Decision Systems Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology CambridgeMA02139 United States

This paper proposes a distributed prescribed-time observer for nonlinear systems representable in a block-triangular observable canonical form. Using a weighted average of neighbor estimates exchanged over a strongly connected digraph, each observer estimates the system state despite limited local sensor measurements. The proposed design guarantees that distributed state estimation errors converge to zero at a user-specified convergence time, irrespective of observers’ initial conditions. To achieve this prescribed-time convergence, distributed observers implement time-varying local output injection gains that asymptotically approach infinity as the prescribed time is approached. The theoretical convergence is rigorously proven and validated through numerical simulations. Copyright © 2025, The Authors. All rights reserved.

关键词： Nonlinear simulations

来源：评论

学校读者我要写书评

暂无评论

Adaptive UAV-Assisted Hierarchical Federated Learning: Optimizing Energy, Latency, and Resilience for Dynamic Smart IoT Networks

arXiv

引用

arXiv 2025年

作者： Yang, Xiaohong Liwang, Minghui Fu, Liqun Su, Yuhan Hosseinalipour, Seyyedali Wang, Xianbin Hong, Yiguang School of Informatics Xiamen University Fujian China Department of Control Science and Engineering The National Key Laboratory of Autonomous Intelligent Unmanned Systems China Frontiers Science Center for Intelligent Autonomous Systems Ministry of Education Tongji University Shanghai China School of Electronic Science and Engineering Xiamen University Xiamen China Department of Electrical Engineering University at Buffalo-SUNY BuffaloNY United States Department of Electrical and Computer Engineering Western University ON Canada

Hierarchical Federated Learning (HFL) introduces intermediate aggregation layers, addressing the limitations of conventional Federated Learning (FL) in geographically dispersed environments with limited communication infrastructure. An application of HFL is in smart IoT systems, such as remote monitoring, disaster response, and battlefield operations, where cellular connectivity is often unreliable or unavailable. In these scenarios, UAVs serve as mobile aggregators, providing connectivity to the terrestrial IoT devices. This paper studies an HFL architecture for energy-constrained UAVs in smart IoT systems, pioneering a solution to minimize global training cost increased caused by UAV disconnection. In light of this, we formulate a joint optimization problem involving learning configuration, bandwidth allocation, and device-to-UAV association, and perform global aggregation in time before UAV drops disconnect and redeployment of UAVs. The problem explicitly accounts for the dynamic nature of IoT devices and their interruptible communications and is unveiled to be NP-hard. To address this, we decompose it into three subproblems. First, we optimize the learning configuration and bandwidth allocation using an augmented Lagrangian function to reduce training costs. Second, we propose a device fitness score, integrating data heterogeneity (via Kullback-Leibler divergence), device-to-UAV distances, and IoT device resources, and develop a twin-delayed deep deterministic policy gradient (TD3)-based algorithm for dynamic device-to-UAV assignment. Third, We introduce a low-complexity two-stage greedy strategy for finding the location of UAVs redeployment and selecting the appropriate global aggregator UAV. Experiments on real-world datasets demonstrate significant cost reductions and robust performance under communication interruptions. © 2025, CC BY-NC-ND.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

DEEPCON: Improving Distributed Deep Learning Model Consistency in Edge-Cloud Environments via Distillation

引用

IEEE Transactions on Cognitive Communications and Networking 2025年

作者： Qian, Bin Qian, Jiaxu Wen, Zhenyu Wu, Di He, Shibo Chen, Jiming Ranjan, Rajiv Zhejiang University State Key Laboratory of Industrial Control Technology Hangzhou China Zhejiang University of Technology Institute of Cyberspace Security College of Information Engineering Hangzhou China University of St Andrews School of Computer Science United Kingdom Newcastle University United Kingdom

In a typical distributed Deep Learning (DL) based application, models are configured differently to meet the requirements of resource constraints. For instance, a large ResNet56 model is deployed on the cloud server while a small lightweight MobileNet model is more suitable for the end-user device with fewer computation resources. However, the heterogeneity of the model architectures and configurations may bring a systemic problem -models may produce different outputs when given the same input. This inconsistency problem may cause severe system failure of prediction agreement inside the application. Current research has not studied the systemic design for efficiently detecting and reducing the inconsistency among models in distributed DL applications. With the increasing scale of distributed DL applications, the challenges of inconsistency mitigation should consider both algorithm and system design. To this end, we design and implement DEEPCON, an adaptive deployment system across the edge-cloud layer with over-the-air model updates. We implement ASRS sampling for efficiently sampling data to reveal the real data distribution as well as model prediction inconsistency. Then, we implement DMML-Par, an asynchronous parallel training algorithm for quickly updating the models and reducing inconsistency. implements over-the-air updates with a set of APIS to enable seamless inconsistency detection and reduction in such deep learning applications. Our experiment results on both vision and language tasks demonstrate that DMML could improve the model consistency up to 4%, 7%, and 13% at CIFAR10/100 and IMDB datasets without sacrificing the accuracy of individual models. We also show that the ASRS sampling can save 90% network bandwidth of data transmission and that DMML-Par is up to 60% faster compared to simple synchronous parallel training. 2015 IEEE.

关键词： Distributed cloud

来源：评论

学校读者我要写书评

暂无评论

Current Challenges and Advancements of Aerial Thermography for Outdoor Structural Health Monitoring: A Review

引用

IEEE Sensors Journal 2025年

作者： Ma, Yue Zeng, Zhi Luo, Zipei Tao, Ning Deng, Lei Tian, Yibin Shenzhen University College of Mechatronics and Control Engineering Shenzhen518060 China Chongqing Normal University College of Computer and Information Science Chongqing401331 China Capital Normal University Department of Physics Key Lab of Terahertz Optoelectronics of the Ministry of Education Beijing100048 China

In the rapid development of outdoor structural health monitoring, infrared thermal imaging and unmanned aerial vehicles (UAVs) have emerged as transformative tools, significantly enhancing maintenance efficiency in expansive outdoor structures. This review highlights the advancements in UAV-assisted thermography, focusing on its impact on the monitoring capabilities of outdoor engineering structures. It introduces UAV-assisted thermographic methods tailored for outdoor environments, explorees current challenges influencing their efficacy, and thoroughly examines their integration with machine learning techniques, demonstrating their applications in identifying structural flaws and deformations. Furthermore, the review emphasizes the transformative potential of 3D thermographic modelling to localize and characterize defects effectively. It contributes actionable insights to optimize UAV-assisted thermographic practices. The findings underscore the importance of integrating UAVs and thermography to overcome existing limitations in outdoor structural health monitoring inspections, setting the stage for a paradigm shift in large-scale outdoor structural condition assessments. © 2001-2012 IEEE.

关键词： Thermography (imaging)

来源：评论

学校读者我要写书评

暂无评论

Boolean matrix compressed sensing

Boolean matrix compressed sensing

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Qiang Liu Mahdi Soleymani Hessam Mahdavifar National Laboratory of Industrial Control Technology Institute of Cyber-Systems and Control Zhejiang University Hangzhou China Halıcıoğlu Data Science Institute University of California San Diego La Jolla CA USA Department of Electrical and Computer Engineering Northeastern University Boston MA USA

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

In real-world datasets, leveraging the low-rank and sparsity properties enables developing efficient algorithms across a diverse array of data-related tasks, including compression, compressed sensing, matrix completion, etc. Notably, these two properties often coexist in certain real-world datasets, especially in Boolean datasets and quantized real-valued datasets. To harness the advantages of low-rank and sparsity simultaneously, we adopt a technique inspired by compressed sensing and Boolean matrix completion. Our approach entails compressing a low-rank sparse Boolean matrix by performing inner product operations with a randomly generated Boolean matrix. We then propose a decoding algorithms based on message-passing techniques to recover the original matrix. Our experiments demonstrate superior recovery performance of our proposed algorithms compared to Boolean matrix completion, with equal measurement requirements.

关键词： Image coding Signal processing algorithms Signal processing Acoustics Sensors Decoding Sparse matrices Speech processing Compressed sensing

来源：评论

学校读者我要写书评

暂无评论

Interpretable deep temporal neural networks for in-situ monitoring under varying conditions in micro-electrical discharge machining

引用

Journal of Manufacturing Systems 2025年 81卷 222-237页

作者： Long Ye Cheng Guo Ming Wu Jun Qian Nan Yu Dominiek Reynaerts Institute of Materials and Processes School of Engineering University of Edinburgh Edinburgh EH9 3FB UK Guangdong Provincial Key Laboratory of Micro/Nano Optomechatronics Engineering College of Mechatronics and Control Engineering Shenzhen University Shenzhen 518060 PR China Micro & Precision Engineering Group Department of Mechanical Engineering KU Leuven Leuven 3001 Belgium Members Flanders Make Leuven Belgium Department of Computer Science KU Leuven Leuven 3001 Belgium

Monitoring is critical enabler of digitalization in modern manufacturing, supporting enhanced process control, quality assurance, and real-time decision-making. By integrating data-driven techniques with the powerful capabilities of deep learning, monitoring systems can efficiently extract valuable insights from complex, high-dimensional time-series data. However, traditional data-driven approaches often lack interpretability, limiting their adoption in industrial applications that demand high reliability and accountability. To address this challenge, this paper proposes an interpretable monitoring framework based on a deep temporal neural network (DTNN). Designed with a modular architecture, the DTNN integrates key components for embedding, temporal feature learning and classification, enabling it to effectively capture complex underlying patterns in temporal process data and overcome the limitations of conventional methods. The DTNN’s capabilities are demonstrated in the context of micro-electrical discharge machining (micro-EDM), a prominent non-traditional machining process known for producing intricate and high-precision components. Through a pulse discrimination task utilizing a large dataset of reliable labels, the DTNN achieves superior classification accuracy under varying processing parameters while providing interpretable insights into discharge phenomena. Furthermore, the DTNN monitoring approach is applied to a deep-hole drilling process in micro-EDM, enabling closed-loop control of discharge status and ensuring long-term process stability. The DTNN’s modular design, interpretability and real-time adaptability underscore its potential for advancing data-driven monitoring systems in digital manufacturing.

关键词：

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：