检索结果-内蒙古大学图书馆

International Conference on computer Vision Workshops (ICCV Workshops)

作者： Jinlin Xiang Eli Shlizerman Department of Electrical & Computer Engineering University of Washington Seattle USA Department of Electrical & Computer Engineering Department of Applied Mathematics University of Washington Seattle USA

When learning multiple tasks in a sequence, deep neural networks tend to loose accuracy on tasks learned in the past while gaining accuracy on the current task. This phenomenon is called catastrophic forgetting. Memory-based Class Incremental Learning (CIL) methods address this problem by re-learning exemplars retained in the memory from previous tasks. However, due to data imbalances between the training data for the current task and the limited exemplars from previous tasks, existing methods struggle to balance the accuracy across all seen tasks. Here, we propose to address data imbalance and in addition to a generic model to learn a set of task-specific parameters. In particular, we propose a novel methodology of Tangent Kernel for Incremental Learning (TKIL) that seeks an equilibrium between current and previous representations. Specifically, TKIL achieves such equilibrium by tuning different task-specific parameters for different tasks with a new Gradient Tangent Kernel (GTK) loss. Therefore, when representing previous tasks, task-specific models are not impacted by the samples of the current task and are able to retain learned representations. As a result, TKIL equally considers the contribution from all task models. The generalized parameters that TKIL obtains allow it to automatically identify which task is being considered and to adapt to it during inference. Extensive experiments on five CIL benchmark datasets with ten incremental learning settings show that TKIL outperforms existing state-of-the-art methods, e.g., achieving 9.4% boost on CIFAR-100 with 25 incremental stages.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Assessing the Impact of Physical Configuration and Lead Time on WRF Forecasting of an Extreme Wind Event

引用

Journal of Meteorological research 2025年第1期39卷 154-171页

作者： Rocío OTERO Matías SUAREZ Edgardo PIEROBON Leandro MATURANO Ignacio MONTAMAT Juan Ezequiel SANCHEZ Lucia SANDALIO Andrés RODRIGUEZ Denis POFFO Hydrometeorology Laboratory Department of ExactPhysical and Natural SciencesNational University of CordobaCórdobaX5016GCAArgentina Institute for Advanced Studies in Engineering and Technology National Scientific and Technical Research CouncilCordobaX5016GCAArgentina Radar and Remote Sensing Laboratory Department of MathematicsAstronomyPhysics and Computer ScienceNational University of CordobaCordobaX5016HUAArgentina

The central region of Argentina is known to be a source of some most extreme weather events in the world,which are partially associated with the passage of cold fronts accompanied often by extreme wind *** may cause severe property damage and even loss of human ***,there is a lack of studies that evaluate the performance of the numerical weather prediction(NWP)models such as weather research and forecasting(WRF)model in anticipating this type of weather in the *** study compares the performance of the operational WRF in Argentina using four combinations of various planetary boundary layer(PBL)and microphysics parameterization schemes under varied lead times in predicting an extreme wind event(gusts>30 m s^(-1))in Central *** results demonstrate that the WRF model is capable of providing an acceptable prediction of wind speed during an extreme *** is found that no single combination outperforms the others,although there is a slight tendency for Combination A,which utilizes the Mellor-Yamada-Janjic(MYJ)parameterization for the PBL and the Eta similarity parameterization for the surface layer,to more accurately capture the extreme wind *** with wind gust observations at five weather stations,the wind gust parameterization predicted the intensity and occurrence time of the peak wind,with an acceptable bias(time of peak<±1 h).Analysis of grid configurations(resolutions of 4 vs 9/3/1 km)revealed that higher resolution does not imply an improvement in the wind gust forecast for this particular *** regard to lead time,a shorter lead time does not necessarily result in more accurate ***,it is beneficial to conduct multiple sensitivity runs in order to obtain and understand the dispersion of forecasted wind speeds.

关键词： extreme wind wind event weather research and forecasting(WRF)model wind gust forecast cold front lead time

来源：评论

学校读者我要写书评

暂无评论

Problems in Video Conferencing-Mediated Teaching: Experiences of Three User Roles of a University in Bangladesh

Problems in Video Conferencing-Mediated Teaching: Experience...

引用

Advances in Computing,Communication, Electrical, and Smart Systems (iCACCESS), International Conference on

作者： Nafees Mansoor Anatte Rozario Muhammad Golam Kibria Md Saifuddin Khalid Department of Computer Science and Engineering University of Liberal Arts Bangladesh Department of Applied Mathematics and Computer Science Technical University of Denmark (DTU) Kongens Lyngby Denmark

Video-conferencing systems (VCS) became a defacto tool for online and hybrid teaching during COVID-19 and the gradual return to physical classrooms. As part of the EdViCon Erasmus+ project, towards defining the scenarios for improving teaching with additional portable tools by university teachers, this study reports users' problems in the context of the University of Liberal Arts Bangladesh (ULAB). Classroom observations, problem-tree analysis, interviews, and focus-group discussions with academics, e-learning and IT consultants, and students provide different perspectives on the central problems, underlying causes, and the effects of VCS use during live teaching sessions. While teachers emphasized the lack of resources and technical barriers, students' problem statement focused on the reduced quality of interactions due to the differences between VCS-mediated and physical classrooms, and the IT & e-learning professionals emphasized technological issues (audio, tone, lack of facial expression etc.), lack of collaboration, and differences in expectations among participants. The central contribution of this paper is three diagrams showing problem tree analysis from the perspectives of the three roles. Each of the causes, effects, and problem statements will be considered as point-of-view statements for generating ideas for supporting hardware and software tools and training on technology and pedagogy,

关键词：

来源：评论

学校读者我要写书评

暂无评论

Implicit Neural Representations and the Algebra of Complex Wavelets

arXiv

引用

arXiv 2023年

作者： Roddenberry, T. Mitchell Saragadam, Vishwanath de Hoop, Maarten V. Baraniuk, Richard G. Department of Electrical and Computer Engineering Rice University United States Department of Computational Applied Mathematics and Operations Research Rice University United States

Implicit neural representations (INRs) have arisen as useful methods for representing signals on Euclidean domains. By parameterizing an image as a multilayer perceptron (MLP) on Euclidean space, INRs effectively represent signals in a way that couples spatial and spectral features of the signal that is not obvious in the usual discrete representation, paving the way for continuous signal processing and machine learning approaches that were not previously possible. Although INRs using sinusoidal activation functions have been studied in terms of Fourier theory, recent works have shown the advantage of using wavelets instead of sinusoids as activation functions, due to their ability to simultaneously localize in both frequency and space. In this work, we approach such INRs and demonstrate how they resolve high-frequency features of signals from coarse approximations done in the first layer of the MLP. This leads to multiple prescriptions for the design of INR architectures, including the use of complex wavelets, decoupling of low and band-pass approximations, and initialization schemes based on the singularities of the desired signal. © 2023, CC BY.

关键词： Chemical activation

来源：评论

学校读者我要写书评

暂无评论

Discrete unified gas kinetic scheme for the solution of electron Boltzmann transport equation with Callaway approximation

引用

Physical Review E 2024年第6期109卷 065310-065310页

作者： Meng Lian Chuang Zhang Zhaoli Guo Jing-Tao Lü Institute for Quantum Science and Engineering and Wuhan National High Magnetic Field Center School of Physics Department of Physics Institute of Interdisciplinary Research for Mathematics and Applied Science

Electrons are the carriers of heat and electricity in materials and exhibit abundant transport phenomena such as ballistic, diffusive, and hydrodynamic behaviors in systems with different sizes. The electron Boltzmann transport equation (eBTE) is a reliable model for describing electron transport, but it is a challenging problem to efficiently obtain the numerical solutions of the eBTE within one unified scheme involving ballistic, hydrodynamics, and/or diffusive regimes. In this work, a discrete unified gas kinetic scheme (DUGKS) in the finite-volume framework is developed based on the eBTE with the Callaway relaxation model for electron transport. By reconstructing the distribution function at the cell interface, the processes of electron drift and scattering are coupled together within a single time step. Numerical tests demonstrate that the DUGKS can be adaptively applied to multiscale electron transport, across different regimes.

关键词： Mesoscopics Physics of computation Transport phenomena Boltzmann theory Hydrodynamics Multiscale modeling

来源：评论

学校读者我要写书评

暂无评论

Real-Time Data-Driven Maintenance Logistics: A Public-Private Collaboration

Real-Time Data-Driven Maintenance Logistics: A Public-Privat...

引用

2024 Commit2Data, C2D 2024

作者： van Jaarsveld, Willem Bliek, Laurens de Weerdt, Mathijs Kapodistria, Stella Pronk, Verus Verleijsdonk, Peter Voorberg, Simon Akçay, Alp da Costa, Paulo Eshuis, Rik Kaymak, Uzay van Houtum, Geert-Jan Verwer, Sicco Zhang, Yingqian Department of Industrial Engineering and Innovation Sciences Eindhoven University of Technology Netherlands Faculty of Electrical Engineering Mathematics and Computer Science Delft University of Technology Netherlands Department of Mathematics and Computer Science Eindhoven University of Technology Netherlands Philips Research Laboratories Eindhoven Netherlands Department of Information Systems Supply Chain Management and Decision Analysis NEOMA Business School Rouen France Etsy Inc. Dublin Ireland Jheronimus Academy of Data Science Eindhoven University of Technology Netherlands

ISBN: (纸本)9783959773515

The project "Real-time data-driven maintenance logistics" was initiated with the purpose of bringing innovations in data-driven decision making to maintenance logistics, by bringing problem owners in the form of three innovative companies together with researchers at two leading knowledge institutions. This paper reviews innovations in three related areas: How the innovations were inspired by practice, how they materialized, and how the results impact practice. © Willem van Jaarsveld, Alp Akçay, Laurens Bliek, Paulo da Costa, Mathijs de Weerdt, Rik Eshuis, Stella Kapodistria, Uzay Kaymak, Verus Pronk, Geert-Jan van Houtum, Peter Verleijsdonk, Sicco Verwer, Simon Voorberg, and Yingqian Zhang.

关键词： Scheduled maintenance

来源：评论

学校读者我要写书评

暂无评论

Securing Neural Network in the Cloud: A Systematic Approach of Cloud Security Based on ANN-SVM Model 4

Securing Neural Network in the Cloud: A Systematic Approach ...

引用

4th International Conference on Sustainable Expert Systems, ICSES 2024

作者： Arunarani, Ar Vijayagopal, V. Kartheesan, L. Dwivedi, Vijay Kumar Rajesh Kumar, A. Nishant, Neerav School of Computing College of Engineering Srm Institute of Science and Technology Department of Computational Intelligence Tamilnadu Chennai India Dr. M.G.R. Educational and Research Institute Faculty of Computer Applications Tamilnadu Chennai India School of Computing Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Department of Computer Science and Engineering Tamilnadu Chennai India Vishwavidyalaya Engineering College Department of Mathematics Chhattisgarh Surguja India N.S.N. College of Engineering and Technology Department of Computer Science and Engineering Tamilnadu Karur India School of Engineering Babu Banarasi Das University Department of Computer Science and Engineering Lucknow India

ISBN: (纸本)9798331540364

Many services have made the transition to the cloud because of the increased reliability and efficiency of cloud computing. Due to their capacity to ease service access while also securing communications in the public network, the three-factor Mutual Authentication and Key Agreement (MAKA) protocols are seeing increased use in multi-server designs. A lack of formal security proof or high communication and computation costs makes the limited three-factor MAKA protocols that are currently available susceptible to a wide range of attacks. The proposed approach consists of three phases, which are preprocessing, feature selection, and model training. Text scheduling and normalization make up preprocessing. In feature selection, the PSO is an optimization method that uses a population-based approach to find optimal solutions. An ANN-SVM was employed throughout the model's training phase. With an average accuracy of 93.30%, the proposed technique surpasses ANN and SVM. © 2024 IEEE.

关键词： Cloud security

来源：评论

学校读者我要写书评

暂无评论

The Design and Fabrication of Smart Surveillance Robot 4

The Design and Fabrication of Smart Surveillance Robot

引用

4th International Conference on Inventive research in Computing Applications, ICIRCA 2022

作者： Thirumaraiselvi, C. Sankara Subramanian, R.S. Janani, R. Kavya Dharshini, M. Keerthi, G. Sri Krishna College of Engineering and Technology Department of Electronics and Communication Coimbatore India Psg Institute of Technology and Applied Research Department of Mathematics Coimbatore India

ISBN: (数字)9781665497077

ISBN: (纸本)9781665497077

Surveillance and monitoring of critical places like international borders, Banks and Industries is currently a difficult endeavor. The security guarding personnel vigilantly monitor the places where suspicious activity can occur, but it is impossible to keep an eye. A robot that automatically detects illegal persons in these areas and reports them to the nearest security control unit is a must-have in this case. Many departments, such as the military, now use robots to perform dangerous tasks that soldiers are unable to perform. In this project, Raspbian spy software based on the Internet of Things (IoT) built-in remote-control system, will provide security and eliminate manual errors. Sensors, Web camera and Raspberry-Pi 4 Model B forms the system of spy robots. Information about the biological PIR sensor identification is transmitted to users via a server, and the web camera captures the object in motion, that is simultaneously inserted into a web page. The user in the control room can use the control buttons to drive the wheels on a web page to control the robot. Object detection sensors manage the robot's motion to avoid collisions. This spy robot monitoring system may be customized for a wide range of businesses, including banks and retail malls. © 2022 IEEE.

关键词： Python

来源：评论

学校读者我要写书评

暂无评论

Fake News Detection Regarding COVID-19 Tweets Using Machine Learning Approaches 8

Fake News Detection Regarding COVID-19 Tweets Using Machine ...

引用

8th IEEE International Conference on Recent Advances and Innovations in engineering, ICRAIE 2023

作者： Hussin, Muhammad Hazimmuddin Mahmud, Yuzi Mohd Hanafiah, Nurul Izzah Azma Nasruddin, Zan Mohd Ariffin, Nor Hapiza Ince, Murat Senel, Fatih Ahmet Universiti Teknologi MARA School of Computing Sciences College of Computing Informatics and Mathematics Shah Alam Malaysia Sohar University Faculty of Business MIS Department Sohar Oman Isparta University of Applied Sciences Department of Computer Technologies Isparta Turkey Suleyman Demiral University Department of Computer Engineering Isparta Turkey

ISBN: (纸本)9798350315516

The pervasiveness of misinformation surrounding the COVID-19 pandemic has garnered heightened attention due to its implications, as a noteworthy proportion of the populace is being exposed to spurious and unsubstantiated narratives concerning the crisis. This research utilizes a dataset sourced from Codalab, comprising 8,560 tweets, with 4,480 labelled as real and 4,080 as fake. The research explores the effectiveness of different machine learning models, including logistic regression (LR), random forest (RF), and deep learning models such as Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (Bi-LSTM). In addition to model comparison, experiments were conducted to analyze the impact of different data splits (70:30, 80:20, and 90:10), batch sizes (16, 32, and 64), and the number of epochs (5, 10, and 15) on model performance. The experiments provided insights into the optimal configurations for the models. The results showcase the model's capabilities, with high accuracy achieved across the different models. Specifically, logistic regression achieved an accuracy of 92%, random forest 91%, Bi-LSTM 93%, and CNN 95%. These findings highlight the potential of deep learning models, particularly CNN, in accurately detecting fake news from COVID-19-related tweets. © 2023 IEEE.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Multivariable Decoupling Control of Civil Turbofan Engines Based on Fully Actuated System Approach

引用

Journal of Systems Science & Complexity 2023年第3期36卷 947-959页

作者： PENG Kai WANG Hongxia ZHANG Huanshui ZHANG Zhaorong YANG Fan School of Power and Energy Northwestern Polytechnical UniversityXi’an 710072China School of Electrical and Automation Engineering Shandong University of Science and TechnologyQingdao 266590China Department of Applied Mathematics The Hong Kong Polytechnic UniversityHung HomKowloonHong KongChina. Xi’an Aeronautics Computing Technique Research Institute AVICXi’an 710068China

Gas turbine engines must be operated by means of control,and how to achieve multivariable control decoupling with aero-engine control constraints is an open thorny issue attracting increasingly more *** paper considers the multivariable decoupling problems of aero-engines by using a compound controller,which originates from the fact that it is impossible to eliminate all the nonlinear dynamics of system to obtain desired constant linear closed-loop system by using full actuated control because of modeling errors and some physical *** controllers are involved in the compound *** is a fully actuated controller and the other is classical feedback *** order to use fully actuated control and maintain the accuracy of engine model,a full state scheduling linear parameter-varying(LPV)modeling method is proposed based on fuzzy neural network *** a general input matrix of the system,its generalized inverse is applied to design fully actuated controller to result in a pseudolinear *** with a feedback controller and control limiter,the control synthesis is *** simulation shows that the proposed method is possessed of a better decoupling and tracking effect compared with traditional control approach.

关键词： Compound control decoupling control fully actuated control turbofan engine

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：