检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Lai, Kuo-Wei Muthukumar, Vidya School of Electrical & Computer Engineering Georgia Institute of Technology United States H. Milton School of Industrial & Systems Engineering Georgia Institute of Technology United States

We provide a unified framework, applicable to a general family of convex losses and across binary and multiclass settings in the overparameterized regime, to approximately characterize the implicit bias of gradient descent in closed form. Specifically, we show that the implicit bias is approximated (but not exactly equal to) the minimum-norm interpolation in high dimensions, which arises from training on the squared loss. In contrast to prior work which was tailored to exponentially-tailed losses and used the intermediate support-vector-machine formulation, our framework directly builds on the primal-dual analysis of [29], allowing us to provide new approximate equivalences for general convex losses through a novel sensitivity analysis. Our framework also recovers existing exact equivalence results for exponentially-tailed losses across binary and multiclass settings. Finally, we provide evidence for the tightness of our techniques, which we use to demonstrate the effect of certain loss functions designed for out-of-distribution problems on the closed-form solution. Copyright © 2023, The Authors. All rights reserved.

关键词： Interpolation

来源：评论

学校读者我要写书评

暂无评论

On the Stability Analysis of Open Federated Learning systems

On the Stability Analysis of Open Federated Learning Systems

引用

American Control Conference (ACC)

作者： Youbang Sun Heshan Fernando Tianyi Chen Shahin Shahrampour the Department of Mechanical and Industrial Engineering Northeastern University Boston Massachusetts USA Department of Electrical Computer and Systems Engineering Rensselaer Polytechnic Institute Troy New York USA

We consider the open federated learning (FL) systems, where clients may join and/or leave the system during the FL process. Given the variability of the number of present clients, convergence to a fixed model cannot be guaranteed in open systems. Instead, we resort to a new performance metric that we term the stability of open FL systems, which quantifies the magnitude of the learned model in open systems. Under the assumption that local clients’ functions are strongly convex and smooth, we theoretically quantify the radius of stability for two FL algorithms, namely local SGD and local Adam. We observe that this radius relies on several key parameters, including the function condition number as well as the variance of the stochastic gradient. Our theoretical results are further verified by numerical simulations on synthetic data.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning Approach for Satellite and Debris Detection: YOLO in Action

A Deep Learning Approach for Satellite and Debris Detection:...

引用

International Conference on computer and Information Technology (ICCIT)

作者： Md Younus Ahamed Md Asif Bin Syed Paroma Chatterjee Al Zadid Sultan Bin Habib Lane Department of Computer Science and Electrical Engineering West Virginia University Morgantown USA Department of Industrial and Management Systems Engineering West Virginia University Morgantown USA

Satellite detection refers to the identification and recognition of satellites in satellite images. This task is crucial for various applications, including space surveillance, satellite tracking, and space debris monitoring. Several studies have been conducted to develop methods and techniques for satellite detection. The intricacy and scale of satellite images make satellite detection incredibly challenging. The process of manual identification and classification necessitates a significant amount of time. Recently, Deep Learning (DL) has emerged as a highly promising approach in computer Vision (CV) tasks. This study evaluates the effectiveness of three widely employed DL-based object detection algorithms, specifically YOLOv4, YOLOv5, and YOLOv8, within the field of satellite detection. The algorithms have undergone training utilizing a dataset of satellite photos comprising both satellites and debris. The evaluation of the algorithm’s performance involved the utilization of broadly accepted metrics, including Precision, Recall, Mean Average Precision (mAP), Box Loss, and Class Loss. The findings indicated that all three algorithms exhibited a high level of accuracy in detecting and precisely determining the location of objects inside satellite data. The performance of YOLOv5 and YOLOv8 surpassed that of YOLOv4 across various measures, with a notable advantage observed in detecting tiny objects. Overall, this study demonstrates the capabilities of DL algorithms in the context of satellite detection and classification techniques.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Optimal Black-Start Restoration Assisted by Mobile Energy Storage

Optimal Black-Start Restoration Assisted by Mobile Energy St...

引用

2022 IEEE Power and Energy Society General Meeting, PESGM 2022

作者： Yip, Joshua Garcia, Manuel Pierre, Brian Kutanoglu, Erhan Santoso, Surya The University of Texas at Austin Electrical and Computer Engineering AustinTX United States Electric Power Systems Research Sandia National Laboratories AlbuquerqueNM United States Operations Research and Industrial Engineering The University of Texas at Austin AustinTX United States

ISBN: (数字)9781665408233

ISBN: (纸本)9781665408233

This paper studies a novel mixed-integer linear programming (MILP) formulation on the application of mobile energy storage (MES) to assist with black-start restoration following the full blackout of an electrical network. By synthesizing techniques in the literature to model generator black start and MES activity, the formulation is the first to integrate the two concepts. Furthermore, it recognizes that the manner in which MES facilitates black-start (BS) restoration may differ depending on what component damages occurred during the event that induced the blackout. Within the IEEE 14-Bus System, testing of the formulation has not only confirmed its efficacy but also underscored circumstances where BS restoration could especially benefit from MES intervention in practice. With an MES sized at 2.59% of total MW generation capacity, in certain damage configuration categories the median load energy unserved is reduced by as much as 45.52 MWh (8.26%), and the median final load supplied is raised by as much as 22.98 MW (10.39%). © 2022 IEEE.

关键词： Outages

来源：评论

学校读者我要写书评

暂无评论

Design of an Electronically Controlled Biquadratic Filter Using Commercial ICs

Design of an Electronically Controlled Biquadratic Filter Us...

引用

13th International electrical engineering Congress, iEECON 2025

作者： Ritnathikul, Somkid Duangmalai, Danupat Kaewdang, Khanittha Kwuanphet, Settasuk Chalermpong, Thanawat Pitaksuttayaprot, Koson Phetchabun Rajabhat University Faculty of Agricultural and IndustrialTechnology Department of Computer Engineering Phetchabun6700 Thailand Ubon Ratchathani University Faculty of Engineering Department of Electrical and Electronic Engineering Ubon Ratchathani34190 Thailand Nakhon Phanom University Faculty of Industrial Technology Department of Electrical Engineering Nakhon Phanom48000 Thailand Phetchabun Rajabhat University Faculty of Agricultural and Industrial Technology Department of Electronics and Automation Systems Engineering Phetchabun6700 Thailand

ISBN: (纸本)9798331543952

This paper presents an electronically controlled biquadratic filter designed for precise frequency selection in applications such as audio processing, telecommunications, and cochlear implants. The design utilizes AD830, AD844, and LM13700 integrated circuits, facilitating high-pass, band-pass, and low-pass responses with improved stability and real-time tunability. PSPICE simulations validate the filter's adaptability and performance, demonstrating its effectiveness in dynamic environments and its suitability for high-frequency applications requiring flexible frequency control. The results indicate a calculated pole frequency of 2.23 MHz, while simulations yield a slightly lower value of 2.0 MHz, affirming the design's practical viability and alignment with theoretical expectations. © 2025 IEEE.

关键词： LSI circuits

来源：评论

学校读者我要写书评

暂无评论

The Impacts of On-Demand Transit Service on Ridership and its Growth

SSRN

引用

SSRN 2024年

作者： Tang, Yili Tasneem, Zaima Zhang, Fangni Department of Civil and Environmental Engineering University of Western Ontario Canada Department of Electrical and Computer Engineering University of Western Ontario Canada Faculty of Engineering and Applied Science University of Regina Canada Department of Industrial and Manufacturing Systems Engineering The University of Hong Kong Hong Kong

On-demand transit service has shown itself to have great prospects in terms of addressing the issues of long-time waiting, connectivity as well as operation efficiency. From the passengers’ perspectives, the service has the potential to improve accessibility with reduced travel times and transfers, whereas transit agencies can benefit from the opportunity of improvement in ridership and operations. As such, this study analyzes the impacts of on-demand transit on ridership and its trend leveraging a combination of transit demand, land use and operation data. The case study is based on a multi-year pilot program of on-demand transit service in the City of Regina, Canada. The impacts of on-demand transit on ridership at stop level as well as the cross-year growth are assessed with difference-in-difference methods for both the overall transit network and different land use zones including residential, commercial and public service areas. Density and distance of various amenities within 500 meters of the transit stops are also considered such as food, school and health facilities. Results prove positive relationships between ridership and on-demand transit services on the entire transit network, while the benefits vary in different land use zones. Results also show increasing impacts of the service on ridership across years which contribute to the ridership growth. The developed method and findings can provide planning and policy guidance for both the existing service evaluation and long-term development for on-demand transit. © 2024, The Authors. All rights reserved.

关键词： Land use

来源：评论

学校读者我要写书评

暂无评论

DMCNET: Data-Driven Multi-Pilot Convolution Neural Network for Mimo-Ofdm Receiver 8

DMCNET: Data-Driven Multi-Pilot Convolution Neural Network f...

引用

8th IEEE International Conference on Network Intelligence and Digital Content, IC-NIDC 2023

作者： Xin, Yutong Peng, Jingyi Lu, Zejian Lee, Yong Yang, Yang China Industrial Control Systems Cyber Emergency Response Team 100029 China China Academic of Electronics and Information Technology 100041 China School of Artificial Intelligence Beijing University of Posts & Telecommunications 100876 China Collage of Electrical and Computer Engineering Sungkyunkwan University 16419 Korea Republic of

ISBN: (纸本)9798350317923

This paper focuses on studying the deep learning (DL) application of neural networks to solve the reception of single-antenna OFDM signals. Specifically, in multi-antenna scenarios, the channel model is more complex compared to single-antenna cases. By leveraging the characteristics of DL, such as automatic learning of parameters using deep neural networks, we treat the reception process of MIMOOFDM signals as a black box and utilize neural networks to accomplish the signal reception task. Moreover, we propose the data-driven multi-pilot convolution neural network for MIMO-OFDM receiver (DMCNet). By incorporating complex convolution and complex fully connected structures, we design a receiver network to recover the transmitted signals from the received signals. We validate the accuracy and robustness of DMCNet under different channel conditions, comparing the bit error rates with different schemes. Additionally, we discuss the factors influencing various channel effects. Experimental results demonstrate that the DL-based reception scheme exhibits promising feasibility in terms of accuracy and interference resistance when compared to traditional approaches. © 2023 IEEE.

关键词： MIMO systems

来源：评论

学校读者我要写书评

暂无评论

Impact of Image Sensor Input Faults on Pruned Neural Networks for Object Detection

Impact of Image Sensor Input Faults on Pruned Neural Network...

引用

IEEE International Symposium on Defect and Fault Tolerance in VLSI systems

作者： Yizhi Chen Wenyao Zhu Dejiu Chen Omar Mohammed Parthib Khound Zhonghai Lu School of Electrical Engineering and Computer Science KTH Royal Institute of Technology Stockholm Sweden School of Industrial Engineering and Management KTH Royal Institute of Technology Stockholm Sweden Chair of Reliability of Technical Systems and Electrical Measurement University of Siegen Siegen Germany

ISBN: (数字)9798350366884

ISBN: (纸本)9798350366891

Object detection is one of the most fundamental problems in computer vision, and image sensors are commonly used for this. In this paper, we present the impact of image sensor faults on pruned neural networks for object detection. We compare the error sensitivities of networks after network slimming, networks after magnitude-based pruning, and native compact models. We also explore different spatial fault types with three intensities. Furthermore, we have developed a temporal error model based on realistic aging image sensor faults. The results illuminate that the performance on clean images is important as the mean Average Precision (mAP) experiences a decrease with an increase in injected faults. Additionally, we demonstrate that the size of the model does not invariably yield a decisive impact on error tolerance when comparing small models such as pruned models and native compact models.

关键词： Image sensors Performance evaluation Fault tolerance Sensitivity Neural networks Fault tolerant systems Object detection Aging Very large scale integration Transformers

来源：评论

学校读者我要写书评

暂无评论

Wireless Power Transmission Using Additively Manufactured Ferrite Pad 1

Wireless Power Transmission Using Additively Manufactured Fe...

引用

1st IEEE International Conference on Smart and Sustainable Developments in electrical engineering, SSDEE 2025

作者： Bhnadari, Shreejan Khanal, Samip Akter, Taslima Li, Frank X Yelamanchi, Bharat Itapu, Srikanth Borra, Vamsi Youngstown State University Electrical and Computer Engineering YoungstownOH United States Chemical Engineering Youngstown State University YoungstownOH United States Industrial and Systems Engineering Youngstown State University YoungstownOH United States Alliance University Engineering and Technology Karnataka Bengaluru India

ISBN: (纸本)9798331542108

Wireless power transfer (WPT) is a promising technology for various purposes, but efficiency remains a key challenge. This research investigates using additively manufactured ferrite pads to enhance wireless power transmission efficiency. By utilizing manganese-zinc ferrite cores with precise 3D-printed geometries, the study aims to improve magnetic flux coupling and reduce energy losses. Using ANSYS Maxwell simulations and experimental setups with AWG 40 Litz wire coils, the research explored how ferrite materials impact power transfer performance. The findings demonstrate the significant improvements: simulations achieved a transfer efficiency of 89% while the experimental data showed output results increasing from 26.27 mV to 40.46 mV when using a ferrite pad. The study addresses critical WPT challenges such as coil misalignment, distance variations, and power density, offering insights into developing more efficient wireless charging solutions and other applications. © 2025 IEEE.

关键词： Electric losses

来源：评论

学校读者我要写书评

暂无评论

A Dual Accelerated Method for Online Stochastic Distributed Averaging: From Consensus to Decentralized Policy Evaluation

arXiv

引用

arXiv 2022年

作者： Zhang, Sheng Pananjady, Ashwin Romberg, Justin Schools of Industrial and Systems Engineering United States Electrical and Computer Engineering Georgia Institute of Technology United States

Motivated by decentralized sensing and policy evaluation problems, we consider a particular type of distributed stochastic optimization problem over a network, called the online stochastic distributed averaging problem. We design a dual-based method for this distributed consensus problem with Polyak–Ruppert averaging and analyze its behavior. We show that the proposed algorithm attains an accelerated deterministic error depending optimally on the condition number of the network, and also that it has an order-optimal stochastic error. This improves on the guarantees of state-of-the-art distributed stochastic optimization algorithms when specialized to this setting, and yields—among other things—corollaries for decentralized policy evaluation. Our proofs rely on explicitly studying the evolution of several relevant linear systems, and may be of independent interest. Numerical experiments are provided, which validate our theoretical results and demonstrate that our approach outperforms existing methods in finite-sample scenarios on several natural network topologies. © 2022, CC BY.

关键词： Linear systems

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：