检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Zong, Weiyu Feng, Mingqian Heyrich, Griffin Chin, Peter Boston University Department of Mathematics and Computer Science United States Johns Hopkins University Department of Applied Mathematics and Statistics United States Dartmouth College Thayer School of Engineering United States

In recent years, deep-learning-based approaches have been introduced to solving time-series forecasting-related problems. These novel methods have demonstrated impressive performance in univariate and low-dimensional multivariate time-series forecasting tasks. However, when these novel methods are used to handle high-dimensional multivariate forecasting problems, their performance is highly restricted by a practical training time and a reasonable GPU memory configuration. In this paper, inspired by a change of basis in the Hilbert space, we propose a flexible data feature extraction technique that excels in high-dimensional multivariate forecasting tasks. Our approach was originally developed for the National Science Foundation (NSF) Algorithms for Threat Detection (ATD) 2022 Challenge. Implemented using the attention mechanism and Convolutional Neural Networks (CNN) architecture, our method demonstrates great performance and compatibility. Our models trained on the GDELT Dataset finished 1st and 2nd places in the ATD sprint series and hold promise for other datasets for time series forecasting. Copyright © 2022, The Authors. All rights reserved.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Advancing Human Fall Detection Through Motion Data from Smart Wearable Sensors and CNN-BiGRU Model

Advancing Human Fall Detection Through Motion Data from Smar...

引用

IEEE International Conference on Software engineering and Service Sciences (ICSESS)

作者： Sakorn Mekruksavanich Wikanda Phaphan Anuchit Jitpattanakul Department of Computer Engineering School of Information and Communication Technology University of Phayao Phayao Thailand Research Group in Statistical Learning and Inference Department of Applied Statistics Faculty of Applied Science King Mongkut's University of Technology North Bangkok Bangkok Thailand Intelligent and Nonlinear Dynamic Innovations Research Center Department of Mathematics Faculty of Applied Science King Mongkut's University of Technology North Bangkok Bangkok Thailand

ISBN: (数字)9798350389166

ISBN: (纸本)9798350389173

Falls pose a significant health risk, especially for older adults, resulting in serious injuries, limited movement, and higher medical expenses. Precise and fast identification of falls is essential for immediate healthcare treatment and avoidance of additional complications. This paper presents a new method to improve the identification of individual falls by using movement inputs from smart wearable sensors and a hybrid deep learning model that combines convolutional neural networks (CNN) and bidirectional gated recurrent units (BiGRU). Intelligent wearable sensors, including accelerometers and gyroscopes, are used to collect real-time movement data from different parts of the human body. The movement signals gathered are subjected to preprocessing and segmentation into windows of a particular length to extract pertinent information. An advanced neural network structure comprising CNN layers for extracting spatial features and BiGRU layers for representing temporal sequences is specifically intended to acquire the distinctive patterns related to fall occurrences efficiently. The proposed CNN-BiGRU model is trained on a varied dataset that includes simulated falls and activities of daily living conducted by several people. The model's effectiveness is assessed using established measures, such as accuracy, sensitivity, specificity, and F1-score. The experimental findings demonstrate that the CNN-BiGRU model has a notable level of accuracy in recognizing instances of falls. Combining in-telligent wearable sensors and advanced deep-learning algorithms creates a dependable and highly effective system for detecting falls. The suggested strategy can significantly enhance the well-being of those prone to falls by offering prompt aid and mitigating the impact of injuries caused by falls.

关键词： Accuracy Sensitivity Neural networks Feature extraction Real-time systems Convolutional neural networks Older adults Wearable sensors Injuries Software engineering

来源：评论

学校读者我要写书评

暂无评论

computer Vision-Driven AI Techniques for Classification of Animal Species

Computer Vision-Driven AI Techniques for Classification of A...

引用

International Conference on Developments in eSystems engineering (DESE)

作者： Anthony Thomas Bacon Abbas Saad Alatrany Luke Topham Hoshang Kolivand Iftikhar Khan Abir Hussain Wasiq Khan School of Computer Science and Mathematics Liverpool John Moores University Liverpool UK School of Pharmacy & Biomolecular Sciences Liverpool John Moores University Liverpool UK Electrical Engineering Department University of Sharjah Sharjah United Arab Emirates

ISBN: (数字)9798350368697

ISBN: (纸本)9798350368703

This study explores the application of deep learning techniques, specifically Convolutional Neural Networks, for the classification of dog breeds from images. By employing varying input image resolutions, the research evaluates the impact of resolution on the accuracy and efficiency of the model. Five experiments were conducted using resolutions of 64, 128, 224, 256, and 512 pixels to assess model performance. The results indicate that an input resolution of 256x256 pixels yields the highest accuracy, achieving 94.74% with an optimal balance between detail and processing complexity. However, certain breeds, such as the American Foxhound and Anatolian Shepherd Dog, exhibited lower classification performance, highlighting the importance of considering breed-specific characteristics in model development. The findings emphasize the critical role of image resolution in training deep learning models and suggest that a 256x256 resolution offers the best trade-off between accuracy and computational efficiency for dog breed classification.

关键词： Deep learning Training Visualization Image resolution Accuracy Computational modeling Data preprocessing Dogs Computational efficiency Optimization

来源：评论

学校读者我要写书评

暂无评论

Numerical Approach of Laser-Induced Spherical Microcavitation Bubbles: Effect of Enthalpy

引用

Bulletin of the Russian Academy of Sciences: Physics 2024年第Suppl 3期88卷 S287-S292页

作者： Ali F. Abu-Bakr Abu-Nab, Ahmed K. AbuShaeer, Zain Department of Mathematics and Computer Science Faculty of Science Menoufia University Shebin El-Koom32511 Egypt Phystech School of Applied Mathematics and Informatics Moscow Institute of Physics and Technology Dolgoprudny Moscow141700 Russia Center for Photonics and 2D Materials Moscow Institute of Physics and Technology Dolgoprudny141700 Russia Department of Basic Science Higher Institute of Engineering and Technology Kafrelsheikh33511 Egypt

Abstract: Over the past decades, interest has increased in studying different techniques to detect the importance of microbubbles in many industrial and medical applications such as mechanisms of hydraulic machinery cavitation erosion, histotripsy, and lithotripsy. The behavior from early irregular oscillations to late regular oscillations of spherical microcavitation bubbles produced by lasers in water was studied by means of plasma photography. Repetition of such delayed oscillations results in quantities of gas/vapour inside the spherical bubbles. Using the modified Gilmore model of cavitation microbubble dynamics, calculations were made for the laser pulse energy, microbubble radius behavior, and microbubble growth rate. Additionally, the pressure evolution inside the bubble, and pressures distribution surrounding of microbubbles in liquid are investigated under the effect of different values of initial cavitation of microbubble radius. The effect of enthalpy is considered into account. The proposed model is solved numerically using numerical techniques as a finite difference method. In addition, the radius of the microbubbles is calculated as a function of time under the influence of the initial radius of these laser-induced microcavitations. Moreover, some physical properties and numerical calculations of the presented mathematical model for cavitation induced by lasers bubbles are deduced, analyzed and investigated. © Pleiades Publishing, Ltd. 2024.

关键词： Cavitation

来源：评论

学校读者我要写书评

暂无评论

Ultra-Low-Power Graphene-Nanoribbon-Based Current-Starved Ring Oscillator

Ultra-Low-Power Graphene-Nanoribbon-Based Current-Starved Ri...

引用

International Conference on Semiconductor

作者： F.-S. Dumitru M. Enachescu A. Antonescu N. Cucu-Laurenciu S. Cotofana Devices Circuits and Architectures Department Faculty of Electronics Telecommunications and Information Technology University Politehnica of Bucharest Bucharest Romania High Energy Physics Department IMAP Radboud University The Netherlands Faculty of Electrical Engineering Mathematics and Computer Science Delft University of Technology The Netherlands

ISBN: (数字)9798350352078

ISBN: (纸本)9798350352085

Identifying methods to further push the boundaries of existing low-power designs has gained new traction, driven by the wide-scale use of large language models. Graphene is well-suited for ultra-low-power nano-electronics due to its exceptional characteristics like ballistic transport, flexibility, and bio-compatibility. In this paper we investigate the feasibility of using ultra-low-voltage GNR structures, in conjunction with power reduction techniques to implement a GNR-based current-starved ring oscillator. By exploiting their low operating voltage and attofarad range intrinsic capacitances we achieve a 1.89 x higher output frequency while simultaneously reducing the power consumption by 553.8 x and achieving a 812 x higher power efficiency.

关键词： Ring oscillators Electric potential Power demand Large language models Graphene Voltage Logic gates FinFETs Capacitance Topology

来源：评论

学校读者我要写书评

暂无评论

Improving the Estimation of Coronary Artery Disease by Classification Machine Learning Algorithm 6

Improving the Estimation of Coronary Artery Disease by Class...

引用

6th IEEE International Conference and Workshop Obuda on electrical and Power engineering, CANDO-EPE 2023

作者： Ghorbani, Hamzeh Krasnikova, Alla Ghorbani, Parvin Ghorbani, Simin Hovhannisyan, Harutyun S. Minasyan, Arsen Minasian, Natali Alvar, Mehdi Ahmadi Stepanyan, Harutyun Azodinia, Mohammazreza Faculty of General Medicine 38a Marshal Babajanyan St. Yerevan0040 Armenia Yerevan State Medical University Faculty of Medicine Department of Cardiology Yerevan0031 Armenia Ahvaz Jundishapur University of Medical Sciences Faculty of Medicine Department of Cardiology Ahvaz Iran Ahvaz Jundishapur University of Medical Sciences Faculty of Nursing Department of Nursing and Midwifery Ahvaz Iran Yerevan State Medical University Department of Internal Disease Propaedeutics Yerevan Armenia Yaroslavl State Medical University Faculty of General Medicine Yaroslavl Russia Shahid Chamran University Faculty of Engineering Department of Computer Engineering Ahwaz Iran Doctoral School of Applied Informatics and Applied Mathematics Obuda University Budapest Hungary

ISBN: (纸本)9798350328752

This research paper aims to predict coronary artery disease (CAD) using data from 350 patients collected at one of the hospitals in Armenia. CAD is a critical parameter which can have a significant impact on patients' life and survival. The study considers several input variables, including level of cholesterol (LOC), patient's age (PA), type of chest pain (TCP), number of arteries blocked (NAB), sex (S), and family history (FH), to make accurate predictions. To achieve this crucial task of CAD prediction, the researchers employed three powerful classification algorithms: Random Forest (RF), Support Vector Machine (SVM), and Logistic Regression (LR). Among these, the Random Forest algorithm stands out for its robustness and numerous advantages, including high accuracy, ability to handle outliers effectively, provision of feature importance insights, and reduced risk of overfitting. The research findings presented in this article demonstrate the impressive performance of the Random Forest algorithm, showcasing an accuracy value of 0.95 and a precision value of 0.94. These results indicate the model's ability to make precise and reliable predictions, essential when dealing with a life-or-death parameter like CAD. By conducting a comparative analysis based on statistical parameters, the researchers establish that Random Forest outperforms both SVM and LR. Thus, the conclusion drawn from the study suggests that the ranking of the algorithms based on their performance is as follows: RF > SVM > LR. © 2023 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Tensor Levenberg-Marquardt Algorithm for Multi-Relational Traffic Prediction

引用

IEEE Transactions on Vehicular Technology 2023年第9期72卷 11275-11290页

作者： Chang, Shih Yu Wu, Hsiao-Chun Kuan, Yen-Cheng Wu, Yiyan San Jose State University Department of Applied Data Science San JoseCA95192 United States Louisiana State University School of Electrical Engineering and Computer Science Baton RougeLA70803 United States Yuan Ze University Innovation Center for AI Applications Chungli32003 Taiwan International College of Semiconductor Technology Hsinchu30010 Taiwan Communications Research Centre Canada OttawaONK2H 8S2 Canada

Traffic prediction serves as a crucial role in intelligent transportation systems for applications including routing planning and traffic control. With the advances of sensor technologies and Internet of Things (IoTs), the exploitation of traffic data motivates us to reformulate the traffic-state prediction problem based on the big-data characteristics. In order to infer multifarious traffic information, the standard one/two-relational traffic data in the vector/matrix form needs to be expanded to multi-relational traffic data in an arbitrary tensor form. However, none of the existing approaches is capable of performing traffic prediction by characterizing and tracking the inherent multi-relational nonlinear dynamics which are often encountered in the realistic time-series analysis. Furthermore, there hardly exist convergence studies on convergence condition and convergence speed pertinent to the existing traffic-prediction methods. In order to combat the aforementioned challenges, we propose a new tensor Levenberg-Marquadt algorithm (TLMA) to conduct the multi-relational nonlinear regression for traffic prediction involving tensorized data. Meanwhile, we also derive the lower-bound of the number of iterations required to make the error-norm below a specified tolerance for global convergence. Besides, we derive the conditions for our proposed TLMA to converge to the optimal solution. The convergence rates with respect to different nonlinear cost functions have been derived theoretically. The computational- and memory-complexities of our proposed new TLMA are studied as well. Finally, numerical experiments are conducted to evaluate the traffic prediction performance of our proposed new TLMA over real traffic data in comparison with other existing multi-relational prediction methods. © 1967-2012 IEEE.

关键词： Tensors

来源：评论

学校读者我要写书评

暂无评论

A Novel Observer-Centric Approach for Detecting Faults in Islanded AC Microgrids with Uncertainties

arXiv

引用

arXiv 2022年

作者： Intriago, Gabriel Intriago, Andres Konstantinou, Charalambos Zhang, Yu The Department of Electrical and Computer Engineering the University of California Santa Cruz United States The Department of Mathematics the Escuela Politécnica Nacional Ecuador The Computer Electrical and Mathematical Sciences and Engineering Division King Ab-dullah University of Science and Technology Saudi Arabia

Fault detection is vital in ensuring AC microgrids’ reliable and resilient operation. Its importance lies in swiftly identifying and isolating faults, preventing cascading failures, and enabling rapid power restoration. This paper proposes a strategy based on observers and residuals for detecting internal faults in grid-forming inverters with power-sharing coordination. The dynamics of the inverters are captured through a nonlinear state space model. The design of our observers and residuals considers H−/H∞ conditions to ensure robustness against disturbances and responsiveness to faults. The proposed design is less restrictive than existing observer-based fault detection schemes by leveraging the properties of quadratic inner-boundedness and one-sided Lipschitz conditions. The internal faults considered in this paper include actuator faults, busbar faults, and inverter bridge faults, which are modeled using vector-matrix representations that modify the state space model of the inverters. One significant advantage of the proposed approach is its cost-effectiveness, as it does not require additional sensors. Experiments are conducted on an islanded AC microgrid with three inductive lines, four inductive loads, and four grid-forming inverters to validate the merits of the proposed fault detection strategy. The results demonstrate that our design outperforms existing methods in the field. © 2022, CC BY.

关键词： State space methods

来源：评论

学校读者我要写书评

暂无评论

Optimal Time for Withdrawal of Voluntary Retirement Scheme with aTime -Varying Threshold 2

Optimal Time for Withdrawal of Voluntary Retirement Scheme w...

引用

2nd International Conference on Innovative Mechanisms for Industry Applications, ICIMIA 2020

作者： Thilaka, B. Theetharappan, Nrithya Sri Venkateswara College of Engineering Department of Applied Mathematics Sriperumbudur Chennai India University of California Los Angeles Department of Electrical and Computer Engineering Los Angeles United States

ISBN: (纸本)9781728141671

Voluntary retirement provides the incentive to the staff, who meet the certain criteria involved in the organization. If an organization feels that it is spending too much on the workforce and that it will still be able to obtain an effective output when the workforce is reduced, it will like to reduce its workforce strength by offering suitable incentives. Whenever there is an economic slowdown, organizations prefer to prune their staff strength as a cost cutting measure and offer voluntary retirement schemes to their personnel both in public sectors and private sectors. This paper considers such cases and the mean optimal withdrawal time for voluntary retirement scheme is determined when the reserve level of the organization's finances reaches a certain threshold level by means of the first-passage-time (FPT) distribution for the reserve to reach the threshold. This threshold level is considered to be varying with time, taking into account various factors like inflation, depreciation etc. © 2020 IEEE.

关键词： Personnel

来源：评论

学校读者我要写书评

暂无评论

Communication Efficient ConFederated Learning: An Event-Triggered SAGA Approach

arXiv

引用

arXiv 2024年

作者： Wang, Bin Fang, Jun Li, Hongbin Eldar, Yonina C. The National Key Laboratory of Wireless Communications University of Electronic Science and Technology of China Chengdu611731 China The Department of Electrical and Computer Engineering Stevens Institute of Technology HobokenNJ07030 United States The Faculty of Mathematics and Computer Science Weizmann Institute of Science Rehovot7610001 Israel

Federated learning (FL) is a machine learning paradigm that targets model training without gathering the local data dispersed over various data sources. Standard FL, which employs a single server, can only support a limited number of users, leading to degraded learning capability. In this work, we consider a multi-server FL framework, referred to as Confederated Learning (CFL), in order to accommodate a larger number of users. A CFL system is composed of multiple networked edge servers, with each server connected to an individual set of users. Decentralized collaboration among servers is leveraged to harness all users’ data for model training. Due to the potentially massive number of users involved, it is crucial to reduce the communication overhead of the CFL system. We propose a stochastic gradient method for distributed learning in the CFL framework. The proposed method incorporates a conditionally-triggered user selection (CTUS) mechanism as the central component to effectively reduce communication overhead. Relying on a delicately designed triggering condition, the CTUS mechanism allows each server to select only a small number of users to upload their gradients, without significantly jeopardizing the convergence performance of the algorithm. Our theoretical analysis reveals that the proposed algorithm enjoys a linear convergence rate. Simulation results show that it achieves substantial improvement over state-of-the-art algorithms in terms of communication efficiency. Copyright © 2024, The Authors. All rights reserved.

关键词： Learning 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：