检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Farajzadeh, Amin Yadav, Animesh Yanikomeroglu, Halim Lab Department of Systems and Computer Engineering Carleton University OttawaONK1S 5B6 Canada School of EECS Ohio University AthensOH45701 United States

The deployment of federated learning (FL) in non-terrestrial networks (NTN) that are supported by high-altitude platform stations (HAPS) offers numerous advantages. Due to its large footprint, it facilitates interaction with a large number of line-of-sight (LoS) ground clients, each possessing diverse datasets along with distinct communication and computational capabilities. The presence of many clients enhances the accuracy of the FL model and speeds up convergence. However, the variety of datasets among these clients poses a significant challenge, as it leads to pervasive non-independent and identically distributed (non-IID) data. The data non-IIDness results in markedly reduced training accuracy and slower convergence rates. To address this issue, we propose a novel weighted attribute-based client selection strategy that leverages multiple user-specific attributes, including historical traffic patterns, instantaneous channel conditions, computational capabilities, and previous-round learning performance. By combining these attributes into a composite score for each user at every FL round and selecting users with higher scores as FL clients, the framework ensures more uniform and representative data distributions, effectively mitigating the adverse effects of non-IID data. Simulation results corroborate the effectiveness of the proposed client selection strategy in enhancing FL model accuracy and convergence rate, as well as reducing training loss, by effectively addressing the critical challenge of data non-IIDness in large-scale FL system implementations. Copyright © 2024, The Authors. All rights reserved.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Linear Projections of Teacher Embeddings for Few-Class Distillation

arXiv

引用

arXiv 2024年

作者： Loo, Noel Iliopoulos, Fotis Hu, Wei Vee, Erik MIT Computer Science Artificial Intelligence Lab United States Google Research United States University of Michigan Department of Computer Science and Engineering United States

Knowledge Distillation (KD) has emerged as a promising approach for transferring knowledge from a larger, more complex teacher model to a smaller student model. Traditionally, KD involves training the student to mimic the teacher's output probabilities, while more advanced techniques have explored guiding the student to adopt the teacher's internal representations. Despite its widespread success, the performance of KD in binary classification and few-class problems has been less satisfactory. This is because the information about the teacher model's generalization patterns scales directly with the number of classes. Moreover, several sophisticated distillation methods may not be universally applicable or effective for data types beyond computer Vision. Consequently, effective distillation techniques remain elusive for a range of key real-world applications, such as sentiment analysis, search query understanding, and advertisement-query relevance assessment. Taking these observations into account, we introduce a novel method for distilling knowledge from the teacher model's representations, which we term Learning Embedding Linear Projections (LELP). Inspired by recent findings about the structure of final-layer representations, LELP works by identifying informative linear subspaces in the teacher's embedding space, and splitting them into pseudo-subclasses. The student model is then trained to replicate these pseudo-subclasses. Our experimental evaluations on large-scale NLP benchmarks like Amazon Reviews and Sentiment140 demonstrate that LELP is consistently competitive with, and typically superior to, existing state-of-the-art distillation algorithms for binary and few-class problems, where most KD methods suffer. Copyright © 2024, The Authors. All rights reserved.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Unleashing Future Donors Navigating Blood Contribution Propensity with Machine Learning Expertise

Unleashing Future Donors Navigating Blood Contribution Prope...

引用

Smart Power Control and Renewable Energy (ICSPCRE), IEEE International Conference on

作者： Kavindra Singh Dhami Raj Gusain Sushant Shekhar Anurag Vidyarth Rishi Prakash R. Gowri Department of Civil Engineering Propagation Research Lab Graphic Era (Deemed to be University) Dehradun India Department of Electronics & Communication Engineering Propagation Research Lab Graphic Era (Deemed to be University) Dehradun India Department of Computer Science Engineering Graphic Era Hill University Dehradun India

ISBN: (数字)9798350377002

ISBN: (纸本)9798350377019

The study used machine learning techniques, which are an essential component of blood donation and health care. Data details of blood donation and service centers have been included in this study. Some of these variables include the number of times a person donates, the frequency of donations, and the time since the last donation. Modern technology machine learning methods analyze and consider key factors in the model. This is valuable information for physicians and blood transfusion centers because here we can predict when a donor may be able to donate blood again. Furthermore, our study aims to examine not only donor behavior but also whether blood donation will provide a continued flow of blood products to health systems in the future. The study of blood donation system is enhanced by this analysis method, which also provides an indication to take care of the health care of the society. We have taken our extensive analysis into account and used cross-validation and performance F1 scores. Donor confidentiality, equity and ethics are equally important to us in the development of our studies. Our approach can play an important role in the technological development sector. Our aim is to draw attention to how donor contact strategies will work, and to improve this, with a view to making the blood supply more reliable.

关键词： Time-frequency analysis Renewable energy sources Ethics Analytical models Navigation Power control Machine learning

来源：评论

学校读者我要写书评

暂无评论

Scale-Aware Super-Resolution Network With Dual Affinity Learning for Lesion Segmentation From Medical Images

引用

IEEE Transactions on Neural Networks and Learning Systems 2024年第6期36卷 10530-10543页

作者： Luyang Luo Yanwen Li Zhizhong Chai Huangjing Lin Pheng-Ann Heng Hao Chen Department of Computer Science and Engineering The Hong Kong University of Science and Technology Sai Kung Hong Kong Department of Biomedical Informatics Harvard University Cambridge MA USA Imsight AI Research Lab Shenzhen China Department of Computer Science and Engineering The Chinese University of Hong Kong Shatin Hong Kong Department of Computer Science and Engineering Department of Chemical and Biological Engineering The Hong Kong University of Science and Technology Hong Kong China HKUST Shenzhen-Hong Kong Collaborative Innovation Research Institute Futian Shenzhen China

Convolutional neural networks (CNNs) have shown remarkable progress in medical image segmentation. However, the lesion segmentation remains a challenge to state-of-the-art CNN-based algorithms due to the variance in scales and shapes. On the one hand, tiny lesions are hard to delineate precisely from the medical images which are often of low resolutions. On the other hand, segmenting large-size lesions requires large receptive fields, which exacerbates the first challenge. In this article, we present a scale-aware super-resolution (SR) network to adaptively segment lesions of various sizes from low-resolution (LR) medical images. Our proposed network contains dual branches to simultaneously conduct lesion mask SR (LMSR) and lesion image SR (LISR). Meanwhile, we introduce scale-aware dilated convolution (SDC) blocks into the multitask decoders to adaptively adjust the receptive fields of the convolutional kernels according to the lesion sizes. To guide the segmentation branch to learn from richer high-resolution (HR) features, we propose a feature affinity (FA) module and a scale affinity (SA) module to enhance the multitask learning of the dual branches. On multiple challenging lesion segmentation datasets, our proposed network achieved consistent improvements compared with other state-of-the-art methods. Code will be available at: https://***/poiuohke/SASR_Net .

关键词： Lesions Image segmentation Feature extraction Medical diagnostic imaging Superresolution Diseases Decoding Computed tomography Technological innovation Skin

来源：评论

学校读者我要写书评

暂无评论

CPL-SLAM: Centralized Collaborative Multi-Robot Visual-Inertial SLAM Using Point-and-Line Features

引用

IEEE Internet of Things Journal 2025年第12期12卷 21866-21875页

作者： Liu, Xin Wen, Shuhuan Liu, Huaping Richard Yu, F. Yanshan University Engineering Research Center The Ministry of Education for Intelligent Control System and Intelligent Equipment Key Lab of Intelligent Rehabilitation and Neuroregulation in Hebei Province Department of Key Lab of Industrial Computer Control Engineering of Hebei Province Qinhuangdao066004 China Tsinghua University Department of Computer Science and Technology Beijing100084 China Shenzhen University College of Computer Science and Software Engineering China Carleton University Canada

Traditional visual-inertial SLAM (Simultaneous Localization and Mapping) systems predominantly rely on feature point matching from a single robot to realize the robot pose estimation and environment map construction. However, in complex scenarios, these traditional systems struggle with issues such as tracking failures due to illumination changes, rapid movements, and low-texture environments, and they perform poorly in terms of mapping efficiency and global consistency. To address these challenges, we propose a centralized collaborative SLAM system that employs both point and line features for tracking in the robot and map fusion in the cloud. The proposed system leverages the fusion of point and line features across all instances in the process, which allows our method to achieve higher localization accuracy in structured, low-texture scenes. With the aid of classifying gravity-aligned vertical lines and spatial parallel lines, the proposed system can deliver faster and more accurate odometry in complex scenes. Furthermore, we developed intra-robot and inter-robot loop closure detection methods based on point and line features, generating a globally consistent sparse point cloud and structured scene map in the cloud. Our method is able to build richer maps while improving accuracy compared to existing methods. Experimental results on public datasets and in real-world environments show that, compared to existing advanced methods, our approach demonstrates better performance. © 2014 IEEE.

关键词： SLAM robotics

来源：评论

学校读者我要写书评

暂无评论

Prototype Development of a Multilevel Solar Dryer for Agriculture

Prototype Development of a Multilevel Solar Dryer for Agricu...

引用

IEEE Region 10 International Conference TENCON

作者： Gian Carlo C. Fernandez Jed Matthew A. San Luis Robert Alfie S. Peña Erees Queen B. Macabebe Department of Electronics Computer and Communications Engineering Ateneo Energy Research Lab School of Science and Engineering Ateneo de Manila University Quezon City Philippines

ISBN: (数字)9798350350821

ISBN: (纸本)9798350350838

Basic food processing in the Philippine agricultural sector relies on drying, which makes food products less perishable. A common practice is traditional sun-drying, which is simple, but exposes the produce to contaminants. This study developed a multilevel solar dryer for agriculture to provide better drying conditions while keeping costs down. A temperature and humidity monitoring system was built into the dryer to control its cooling fans and the humidity inside. The dryer demonstrated better drying conditions with an average drying temperature of 47.38 °C compared to the traditional sun-drying's 39.38 °C. In addition, the solar dryer's inside humidity was also better with a recorded average of 31.29% RH compared to the traditional setup's 39.66% RH. The dryer successfully dried a hybrid of Capsicum frutescens and Capsicum annuum or siling tingala, as it is locally known, at a drying rate faster than the traditional sun-drying setup. Overall, the multilevel solar dryer was able to provide better drying conditions, a faster drying rate, and a protected enclosed space for drying food products.

关键词： Temperature sensors Temperature measurement Prototypes Humidity Agriculture Food products Sun Monitoring Optimization IEEE Regions

来源：评论

学校读者我要写书评

暂无评论

Adaptive PV Systems: Using Genetic Algorithm and Random Forest Classifier to Mitigate the Effects of Partial Shading on Solar Photovoltaic Systems

Adaptive PV Systems: Using Genetic Algorithm and Random Fore...

引用

IEEE Region 10 International Conference TENCON

作者： Ivan Perry M. Cabanayan Adrian Enrico D. Rodriguez Robert Alfie S. Peña Erees Queen B. Macabebe Department of Electronics Computer and Communications Engineering Ateneo Energy Research Lab School of Science and Engineering Ateneo de Manila University Quezon City Philippines

ISBN: (数字)9798350350821

ISBN: (纸本)9798350350838

Mismatch losses due to partial shading of solar photovoltaic (PV) arrays can limit their generated power. One method to improve their power output is to electrically reconfigure the array based on the principle of row irradiance equalization. Shaded panels in an array are identified to determine the array's optimal configuration and maximize its power output. In this study, a random forest classification model was developed to identify the partial shading pattern incident on an array. Features of the model included the individual panel currents and the array irradiance of preset 20 shading patterns. The data used to train the model was obtained using an off-grid $3\times 3$ array composed of nine 10-W PV modules in total-cross-tied (TCT) topology. The model achieved a mean accuracy of 90% using the testing data and 98% using 5-fold cross-validation. This was integrated with a genetic algorithm-based irradiance equalization heuristic, which takes a predicted shading pattern as input and outputs an optimized configuration. For each shading pattern, the suggested reconfiguration was recorded and its corresponding power change was calculated. Power recovery ranged from −15.70% to 40%. The $3\times 3$ PV array reconfigured successfully to the optimum configuration.

关键词： Photovoltaic systems Accuracy Data models Topology Pattern recognition Arrays Random forests Genetic algorithms Testing IEEE Regions

来源：评论

学校读者我要写书评

暂无评论

Efficient Load Frequency Regulation of Islanded Microgrids with Fractional-Order PI Controller

Efficient Load Frequency Regulation of Islanded Microgrids w...

引用

International Conference on Power Electronic Drives

作者： T. K. Bashishtha V. P. Singh Akhilesh Mathur Krishna Murari Mrityunjay Kumar Mishra Ameena Saad Al-Sumaiti Electrical Engineering M. National Institute of Technology Jaipur RJ India Dept. of Electrical Engineering and Computer Science The University of Toledo Ohio USA EECS Department Smart OR Lab APEC Khalifa University Abu Dhabi UAE

ISBN: (数字)9798350372472

ISBN: (纸本)9798350372489

This article focuses on load frequency regulation of islanded microgrid $(I\mu \mathbf{G})$ , which consists of various distributed generation sources like photovoltaic cells, wind turbines, microturbine, biodiesel engine generators, and biogas turbine generators with energy storage systems such as the battery, flywheel, ultracapacitor, and electric vehicles. To eliminate undesirable frequency deviations and errors of $(I\mu \mathbf{G})$ , a fractional-order proportional-integral (FOPI) controller is utilized. Fine-tuning of FOPI controller gains is performed through a rule-based method. To prove the efficacy of FOPI controller in stabilizing frequency response, various simulation-based studies are conducted and compared. Moreover, comparative frequency response, with and without controller and transient specifications, underscore the importance of FOPI controller in stabilizing microgrid frequency.

关键词： Photovoltaic cells Microgrids Supercapacitors Generators Frequency response Wind turbines Flywheels Transient analysis Frequency control Load modeling

来源：评论

学校读者我要写书评

暂无评论

Secure Communication for MUM-T: A Blockchain and Lightweight Cryptography Framework

Secure Communication for MUM-T: A Blockchain and Lightweight...

引用

Security of Information and Networks (SIN), International Conference on

作者： Halimcan Yaşar Şerif Bahtiyar Department of Computer Engineering Cyber Security and Privacy Research Lab SPFLab Istanbul Technical University Maslak Istanbul Türkiye

ISBN: (数字)9798331509736

ISBN: (纸本)9798331509743

Manned-Unmanned Teaming (MUM-T) systems integrate manned aircraft and unmanned aerial vehicles (UAVs) to enhance mission effectiveness, allowing a single pilot to coordinate multiple UAVs for tasks like reconnaissance, communication, and targeting. However, the complexity and operational demands of MUM-T systems introduce significant security challenges, particularly for mission-critical data integrity and real-time communication. In this paper, we propose a new framework for adaptive blockchain cryptography that combines Proof of Authority (PoA)-based blockchain with XOR-based lightweight authentication. The blockchain component, with the manned aircraft that serves as the sole validator, ensures tamper-resistant logging of key mission data. Additionally, it supports accountability and traceability through an efficient PoA consensus algorithm. In parallel, XOR-based lightweight authentication secures control and telemetry signals with minimal computational overhead that enables low-latency and a real-time communication. Analyses results show that the proposed framework achieves a better transaction throughput with acceptable latency, which meets the stringent security and performance requirements of MUM-T operations. The proposed framework offers a scalable and resilient solution for secure communications in complex military environments.

关键词： Military computing Mission critical systems Authentication Reconnaissance Throughput Real-time systems Blockchains Telemetry Aircraft Low latency communication

来源：评论

学校读者我要写书评

暂无评论

The Impact of Forced Working-From-Home on Code Technical Debt: An Industrial Case Study 48

The Impact of Forced Working-From-Home on Code Technical Deb...

引用

48th Euromicro Conference on Software engineering and Advanced Applications, SEAA 2022

作者： Zabardast, Ehsan Gonzalez-Huerta, Javier Palma, Francis Blekinge Institute of Technology Software Engineering Research Lab SERL Sweden Karlskrona Sweden Linnaeus University Department of Computer Science and Media Technology Kalmar Sweden

ISBN: (纸本)9781665461528

Background: The COVID-19 outbreak interrupted regular activities for over a year in many countries and resulted in a radical change in ways of working for software development companies, i.e., most software development companies switched to a forced Working-From-Home (WFH) mode. Aim: Although several studies have analysed different aspects of forced WFH mode, it is unknown whether and to what extent WFH impacted the accumulation of technical debt (TD) when developers have different ways to coordinate and communicate with peers. Method: Using the year 2019 as a baseline, we carried out an industrial case study to analyse the evolution of TD in five components that are part of a large project while WFH. As part of the data collection, we carried out a focus group with developers to explain the different patterns observed from the quantitative data analysis. Results: TD accumulated at a slower pace during WFH as compared with the working-from-office period in four components out of five. These differences were found to be statistically significant. Through a focus group, we have identified different factors that might explain the changes in TD accumulation. One of these factors is responsibility diffusion which seems to explain why TD grows faster during the WFH period in one of the components. Conclusion: The results suggest that when the ways of working change, the change between working from office and working from home does not result in an increased accumulation of TD. © 2022 IEEE.

关键词： COVID-19

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：