检索结果-内蒙古大学图书馆

2023 IEEE International Symposium on Information Theory, ISIT 2023

作者： Krishnan, M. Nikhil Thomas, Anoop Sasidharan, Birenjith Iit Palakkad Department of Electrical Engineering India Iit Bhubaneswar School of Electrical Sciences India Monash University Department of Electrical and Computer Systems Engineering Australia

ISBN: (纸本)9781665475549

The growing privacy concerns and the communication costs associated with transmitting raw data have resulted in techniques like federated learning, where the machine learning models are trained at the edge nodes, and the parameter updates are shared with a central server. Because communications from the edge nodes are often unreliable, a hierarchical setup involving intermediate helper nodes is considered. The communication links between the edges and the helper nodes are error-prone and are modeled as straggling/failing links. To overcome the issue of link failures, coding techniques are proposed. The edge nodes communicate encoded versions of the model updates to the helper nodes, which pass them on to the master after suitable aggregation. The primary work in this area uses repetition codes and Maximum Distance Separable (MDS) codes at the edge nodes to arrive at the Aligned Repetition Coding (ARC) and Aligned MDS Coding (AMC) schemes, respectively. We propose using vector codes, specifically a family of layered MDS codes parameterized by a variable ?, at the edge nodes. For the proposed family of codes, suitable aggregation strategies at the helper nodes are also developed. At the extreme values of ?, our scheme matches the communication costs incurred by the ARC and AMC schemes, resulting in a graceful transition between these schemes. © 2023 IEEE.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

A Framework of Deep Optimal Features Selection for Apple Leaf Diseases Recognition

引用

computers, Materials & Continua 2023年第4期75卷 697-714页

作者： Samra Rehman Muhammad Attique Khan Majed Alhaisoni Ammar Armghan Usman Tariq Fayadh Alenezi Ye Jin Kim Byoungchol Chang Department of Computer Science HITEC UniversityTaxilaPakistan Computer Sciences Department College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityRiyadh11671Saudi Arabia Department of Electrical Engineering College of EngineeringJouf UniversitySakakahSaudi Arabia Department of Management Information Systems CoBAPrince Sattam bin Abdulaziz UniversitySaudi Arabia Department of Computer Science Hanyang UniversitySeoul04763Korea Center for Computational Social Science Hanyang UniversitySeoul04763Korea

Identifying fruit disease manually is time-consuming, expertrequired,and expensive;thus, a computer-based automated system is widelyrequired. Fruit diseases affect not only the quality but also the *** a result, it is possible to detect the disease early on and cure the fruitsusing computer-based techniques. However, computer-based methods faceseveral challenges, including low contrast, a lack of dataset for training amodel, and inappropriate feature extraction for final classification. In thispaper, we proposed an automated framework for detecting apple fruit leafdiseases usingCNNand a hybrid optimization algorithm. Data augmentationis performed initially to balance the selected apple dataset. After that, twopre-trained deep models are fine-tuning and trained using transfer ***, a fusion technique is proposed named Parallel Correlation Threshold(PCT). The fused feature vector is optimized in the next step using a hybridoptimization algorithm. The selected features are finally classified usingmachine learning algorithms. Four different experiments have been carriedout on the augmented Plant Village dataset and yielded the best accuracy of99.8%. The accuracy of the proposed framework is also compared to that ofseveral neural nets, and it outperforms them all.

关键词： Convolutional neural networks deep learning features fusion features optimization classification

来源：评论

学校读者我要写书评

暂无评论

Enhancing Energy Efficiency and Sustainability with Hybrid AC/DC Microgrids for Energy Transition 3

Enhancing Energy Efficiency and Sustainability with Hybrid A...

引用

3rd International Conference on Energy Transition in the Mediterranean Area, SyNERGY MED 2024

作者： Charalambous, Chrysanthos Polycarpou, Alexis Efthymiou, Venizelos Georghiou, George E. University of Cyprus PV-Technology Laboratory Department of Electrical and Computer Engineering Nicosia Cyprus PHAETHON Centre of Excellence for Intelligent Efficient and Sustainable Energy Solutions Nicosia Cyprus Frederick University Department of Electrical Computer Engineering and Informatics Nicosia Cyprus

ISBN: (纸本)9798350375923

The transition to a lower-carbon energy system across Europe brings numerous opportunities and challenges, necessitating advancements in sustainable and eco-friendly energy solutions. This study examines the design and operation of distribution networks and microgrids, particularly emphasizing the integration of hybrid AC/DC distribution systems. The primary goals of this integration are to enhance overall system performance and reduce operating expenses. Utilizing DIgSILENT Power Factory software, the research simulates a microgrid at the FOSS Research Center for Sustainable Energy at the University of Cyprus, incorporating real load and PV generation data. In order to reduce the amount of energy imported from the main grid, this approach evaluates various combinations of AC and DC loads, power production units, and energy storage solutions. The results show that the most optimal structure, with 65% DC loads and 35% AC loads, maximizes efficiency as well as minimizes running expenses. The analysis highlights a notable decrease in imported energy, which dropped from 60.3 kWh/day at full AC loads to 51.6 kWh/day at the optimal DC ratio. Finally, the design and implementation of effective, sustainable energy systems that are in line with changing European energy policy and technology breakthroughs will be significantly impacted by these outcomes. © 2024 IEEE.

关键词： Energy transition

来源：评论

学校读者我要写书评

暂无评论

OLLIE: Imitation Learning from Offline Pretraining to Online Finetuning 41

OLLIE: Imitation Learning from Offline Pretraining to Online...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Yue, Sheng Hua, Xingyuan Ren, Ju Lin, Sen Zhang, Junshan Zhang, Yaoxue Department of Computer Science and Technology Tsinghua University Beijing China Zhongguancun Laboratory Beijing China Department of Computer Science University of Houston Texas United States Department of Electrical and Computer Engineering University of California Davis United States

In this paper, we study offline-to-online Imitation Learning (IL) that pretrains an imitation policy from static demonstration data, followed by fast finetuning with minimal environmental interaction. We find the naïve combination of existing offline IL and online IL methods tends to behave poorly in this context, because the initial discriminator (often used in online IL) operates randomly and discordantly against the policy initialization, leading to misguided policy optimization and unlearning of pretraining knowledge. To overcome this challenge, we propose a principled offline-to-online IL method, named OLLIE, that simultaneously learns a near-expert policy initialization along with an aligned discriminator initialization, which can be seamlessly integrated into online IL, achieving smooth and fast finetuning. Empirically, OLLIE consistently and significantly outperforms the baseline methods in 20 challenging tasks, from continuous control to vision-based domains, in terms of performance, demonstration efficiency, and convergence speed. This work may serve as a foundation for further exploration of pretraining and finetuning in the context of IL. Copyright 2024 by the author(s)

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Uncovering Spatiotemporal Differences in Cortical Activity Corresponding to two Tasks Using Data-Driven ADASSO Algorithm

Uncovering Spatiotemporal Differences in Cortical Activity C...

引用

IEEE International Symposium on Biomedical Imaging

作者： Siwei Mai Ali E. Haddad Laleh Najafizadeh Department of Electrical and Computer Engineering Integrated Systems and NeuroImaging Laboratory Rutgers University Department of Computer Engineering University of Basrah

ISBN: (数字)9798331520526

ISBN: (纸本)9798331520533

Identifying spatiotemporal differences in brain functional dynamics corresponding to two tasks is critical for under-standing how specific neural processes contribute to distinct tasks or cognitive functions. Traditional methods rely on imposing assumptions and limits on the location and timing of activities, while machine-learning-based methods generally lack offering interpretable insights. This highlights the need for new data-driven approaches to capture spatial and temporal differences in brain activity between two tasks, while also providing interpretable explanations of the neural processes underlying these differences. In this work, we formulate the problem of finding the spatial and temporal differences in the dynamics of brain function corresponding to two motor imagery (MI) tasks (left hand movement vs right hand movement) as a discriminative discrete basis problem (DDBP). We apply the data-driven asymmetric discriminative associative algorithm (ADASSO) to EEG data collected during these tasks to uncover the key functional components of the brain's functional dynamics that differentiate between them. Results suggest that hand movements are strongly associated with high-confidence activation in the motor cortex, verifying the effectiveness of the ADASSO algorithm in identifying the location and timing of cortical activities that distinguish between the two task classes.

关键词： Hands Brain Heuristic algorithms Motors Electroencephalography Brain-computer interfaces Spatiotemporal phenomena Timing Biomedical imaging

来源：评论

学校读者我要写书评

暂无评论

Parasitic Capacitance Analysis of Integrated Transformers with Parallel Windings

Parasitic Capacitance Analysis of Integrated Transformers wi...

引用

2024 IEEE Energy Conversion Congress and Exposition, ECCE 2024

作者： Yuan, Tianlong Jin, Feng Li, Qiang Virginia Polytechnic and State University Center for Power Electronics Systems Bradley Department of Electrical and Computer Engineering Blacksburg United States

ISBN: (纸本)9798350376067

The electric transportation market has experienced rapid growth in recent years, driven by increasing environmental concerns. Resonant converters play a crucial role in this sector due to their ability to provide isolation and regulation functions with high efficiency. A resonant inductance embedded transformer is a critical component of the LLC resonant converter, significantly influencing the power density of the entire system, particularly in high-power applications. To achieve higher power density, the operating frequency often exceeds hundreds of kilohertz. At such high frequencies, parasitic capacitance cannot be ignored, as it may cause electromagnetic interference (EMI) issues. In high-power applications, integrated transformers with parallel windings are commonly used, adding complexity to the system. Measuring the parasitic capacitance of these transformers becomes increasingly challenging. This paper presents various measurement methods for an integrated transformer with parallel windings in a 30 kW LLC resonant converter. The results obtained from different measurement techniques were consistent with each other and with simulation results, validating the accuracy of the methods employed. © 2024 IEEE.

关键词： Resonant converters

来源：评论

学校读者我要写书评

暂无评论

InGaPχ^((2))integrated photonics platform for broadband,ultra-efficient nonlinear conversion and entangled photon generation

引用

Light(Science & Applications) 2024年第12期13卷 3090-3098页

作者： Joshua Akin Yunlei Zhao Yuvraj Misra A.K.M.Naziul Haque Kejie Fang Holonyak Micro and Nanotechnology Laboratory and Department of Electrical and Computer Engineering University of Illinois at Urbana-ChampaignUrbanaIL 61801USA Illinois Quantum Information Science and Technology Center University of Illinois at Urbana-ChampaignUrbanaIL 61801USA

Nonlinear optics plays an important role in many areas of science and *** advance of nonlinear optics is empowered by the discovery and utilization of materials with growing optical *** we demonstrate an indium gallium phosphide(InGaP)integrated photonics platform for broadband,ultra-efficient second-order nonlinear *** InGaP nanophotonic waveguide enables second-harmonic generation with a normalized efficiency of 128,000%/W/cm^(2)at 1.55μm pump wavelength,nearly two orders of magnitude higher than the state of the art in the telecommunication C ***,we realize an ultra-bright,broadband time-energy entangled photon source with a pair generation rate of 97 GHz/mW and a bandwidth of 115 nm centered at the telecommunication C *** InGaP entangled photon source shows high coincidence-to-accidental counts ratio CAR>10^(4)and two-photon interference visibility>98%.The InGaP second-order nonlinear photonics platform will have wide-ranging implications for non-classical light generation,optical signal processing,and quantum networking.

关键词： nonlinear optics coincidence

来源：评论

学校读者我要写书评

暂无评论

A variable structure model of neurons’ application in computer vision

引用

Multimedia Tools and Applications 2025年 1-11页

作者： Sridi, Alaeddine Attia, Mouna Bouallegue, Kais Electronics and Micro-Electronics Laboratory Faculty of Sciences of Monastir University of Monastir Monastir Tunisia The Higher Institute of Computer Science and Communication Technologies of Hammam Sousse University of Sousse Sousse Tunisia Laboratory of Computer Science Parallelism and Intelligent Communicating Systems Faculty of Sciences of Tunisia University of Tunis ELManar Tunis Tunisia Department of Electrical Engineering Higher Institute of Applied Sciences and Technology of Sousse University of Sousse Sousse Tunisia

The neuron doctrine defines the neuron as the basic unit of the nervous system, which drives the dynamic behavior of our organs. This has led to neurons becoming the focus of modern neuroscience research and to the rise of neurocomputing, in which we try to digitally simulate human behavior. The goal being to better understand the human nervous system and gain more control over neuron-related diseases, helping doctors establish faster and more reliable medical diagnoses. In this context, we aim to simulate eye diseases and vision problems by convolving input images using a new class of neural networks based on a variable structure model of neurons presented in [1]. In this work, we focus on recreating the common convolution filters as a first step. This is the first time in neurocomputing literature that dendritic neurons have been used for dynamic convolution kernels generation and eye diseases simulation. We start with a presentation of our upgrades to the model’s flexibility and accuracy. Then, we propose a novel kernel generator algorithm to convolve the input images. Finally, we show its promising results in creating the outline, pixelization, and night convolution kernels, as well as control some metrics, including brightness and data retention. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Neurons

来源：评论

学校读者我要写书评

暂无评论

Adaptive Robust Control Integrated With Gaussian Processes for Quadrotors: Enhanced Accuracy, Fault Tolerance and Anti-Disturbance

引用

IEEE Transactions on systems, Man, and Cybernetics: systems 2025年第5期55卷 3235-3248页

作者： Liang, Weisheng Amer, Abdelhakim Mehndiratta, Mohit Chen, Zheng Yao, Bin Kayacan, Erdal Zhejiang University State Key Laboratory of Fluid Power and Mechatronic Systems Hangzhou310027 China Aarhus University Artificial Intelligence in Robotics Laboratory Department of Electrical and Computer Engineering Aarhus8000 Denmark GIM Robotics Espoo02650 Finland Zhejiang University Ocean College Ocean Research Center of Zhoushan Zhoushan316021 China Purdue University School of Mechanical Engineering West Lafayette47907 United States Department of Electrical Engineering and Information Technology Paderborn33098 Germany

With increasingly challenging applications for quadrotors, higher requirements are emerging for tracking accuracy and safety. While high accuracy is a prerequisite for complex tasks, safety is ensured through tolerance to actuator faults and resistance to external disturbances. In this article, adaptive robust control (ARC) integrated with Gaussian processes (GPs), i.e., ARC-GP, is proposed to achieve enhanced accuracy, fault tolerance, and anti-disturbance. These three requirements are interrelated and affected by uncertainties. The primary idea of this article is to categorize uncertainties into parametric and nonparametric types, which are then addressed through parameter adaptation and GP, respectively. First, a detailed dynamic model is established, including actuator models that reflect different types of faults corresponding to changes in different physical parameters. Then, parameter adaptation is designed, with direct and indirect methods adopted for different parameters. In particular, the actuator parameters are effectively estimated to achieve targeted fault compensation. Regarding GP for nonparametric uncertainties, its model parameters are also updated via parameter adaptation. The GP thereby also learns parameter estimation errors along with external disturbances. Accordingly, ARC controllers are designed, for which robust feedback terms are constructed to further mitigate uncertainties on the basis of the covariances predicted by GP. The experiments demonstrate that the proposed ARC-GP can actively tolerate various types of actuator faults and better resist wind disturbances. © 2013 IEEE.

关键词： Robust control

来源：评论

学校读者我要写书评

暂无评论

Wheat Field Fire Smoke Detection from UAV Images using CNN-CBAM 2

Wheat Field Fire Smoke Detection from UAV Images using CNN-C...

引用

2nd International Conference on Artificial Intelligence, Blockchain, and Internet of Things, AIBThings 2024

作者： Alsalem, Ahmed Zohdy, Mohamed Oakland University Rochester Department of Industrial and Systems Engineering MI United States Oakland University Rochester Department of Electrical and Computer Engineering MI United States

ISBN: (纸本)9798331529529

Wheat field fires pose a significant threat to wildlife habitats, agricultural resources, and especially farmer incomes. The frequency of these fires has increased noticeably in recent years, likely due to human activity in agricultural areas and the exacerbating effects of climate change. Rapid action is crucial once smoke is detected, as it marks the first step in early fire detection. To address this challenge, we propose an enhanced CNN model specifically designed to identify smoke in wheat field fires. As part of this approach, a dataset of 3900 UAV images was compiled to depict smoke from wheat field fires. The proposed work also integrates the CBAM into the CNN to improve the model's ability to extract features for smoke patch characteristics of wheat field fires. Furthermore, the CNN architecture includes decoupled heads to extract relevant data from various sources. Testing on our wheat field fire smoke dataset has demonstrated the effectiveness of this proposed method, achieving an accuracy of up to 0.91 and outperforming the previous model. © 2024 IEEE.

关键词： Unmanned aerial vehicles (UAV)

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：