检索结果-内蒙古大学图书馆

23rd IEEE International Conference on Machine Learning and Applications, ICMLA 2024

作者： Kalapos, András Gyires-Tóth, Bálint Budapest University of Technology and Economics Faculty of Electrical Engineering and Informatics Department of Telecommunications and Artificial Intelligence Muegyetem rkp. 3 BudapestH-1111 Hungary

ISBN: (纸本)9798350374889

Self-supervised learning (SSL) has been shown to be a powerful approach for learning visual representations. In this study, we propose incorporating ZCA whitening as the final layer of the encoder in self-supervised learning to enhance the quality of learned features by normalizing and decorrelating them. Although whitening has been utilized in SSL in previous works, its potential to universally improve any SSL model has not been explored. We demonstrate that adding whitening as the last layer of SSL pretrained encoders is independent of the self-supervised learning method and encoder architecture, thus improving performance for a wide range of SSL methods across multiple encoder architectures and datasets. Our experiments show that whitening is capable of improving linear and k-NN probing accuracy by 1-5 percentage points. Additionally, we incorporate metrics that allow for a comprehensive analysis of the learned features, provide insights into the quality of the representations, and help identify collapse patterns. © 2024 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Clutter Removal in Ground Penetrating Radar by Learned RPCA 46

Clutter Removal in Ground Penetrating Radar by Learned RPCA

引用

46th International Conference on telecommunications and Signal Processing, TSP 2023

作者： Ozgul, Samet Erer, Isin Istanbul Technical University Institute of Informatics Istanbul Turkey Istanbul Technical University Electronics and Telecommunications Engineering Department Istanbul Turkey

ISBN: (纸本)9798350303964

Clutter suppression presents crucial importance for Ground Penetrating Radar (GPR) images since clutter decreases considerably target detection rates. Robust Principal Component Analysis (RPCA) is widely used to remove clutter. However, RPCA requires sequential singular value decomposition (SVD) operations in each iteration, and thus computational cost and run-time increase. Also, the hyperparameter should be set manually. In this paper we propose to use unfolding techniques by converting each iteration to a single layer of the network and train the resulting Convolutional Neural Network (CNN) structure to learn the separation of GPR images into clutter and target components. The proposed method is compared to SVD, traditional RPCA , SVD free RNMF and learning based RAE. The recently introduced public hybrid dataset is used for training. The visual and quantitative results validate a performance which approximates RPCA while outperforming RAE with running times less than in any of the existing methods. © 2023 IEEE.

关键词： Singular value decomposition

来源：评论

学校读者我要写书评

暂无评论

PROJECT-BASED LEARNING FOR engineering STUDENTS IN THE CONTEXT OF INDUSTRY 4.0: APPLICATION TO AUTOMOTIVE ASSEMBLY SYSTEM 24

PROJECT-BASED LEARNING FOR ENGINEERING STUDENTS IN THE CONTE...

引用

24th International Conference on engineering Design, ICED 2023

作者： Al Khatib, Ahmad Malhaire, Jean-Marie Dauvé, Stéphane Fougères, Alain-Jérôme Ecam Rennes - Louis de Broglie Bruz France Dept. of Mechanical and Industrial Engineering Dept. of Informatics & Telecommunications

Fourth industrial revolution called Industry 4.0 has radically transformed production systems in manufacturing companies by the integration of emerging technologies. However, manufacturers must overcome several barriers, such as the lack of qualified talent to develop and manage various high-technology systems. Assembly system design aims to define proper assembly line configurations with the optimal performances to overcome increased competitiveness in the market. Nowadays, assembly system design should consider industry 4.0 concepts integration beyond traditional aspects like system balancing and sequencing. In this paper, we introduce a project-based learning approach to teach engineering students assembly system design taking into account industry 4.0 dimension. This project is carried out in collaboration with an industrial partner to design and implement car doors assembly line. The project demonstrated students interest and prepared them better for industry 4.0 era. © The Author(s), 2023. Published by Cambridge University Press.

关键词： Industry 4.0

来源：评论

学校读者我要写书评

暂无评论

Exploring Ensemble Machine Learning Models for Attention and Memory Assesment 46

Exploring Ensemble Machine Learning Models for Attention and...

引用

46th International Conference on telecommunications and Signal Processing, TSP 2023

作者： Gramouseni, Foteini Prapas, Georgios Bellos, Christos Angelidis, Pantelis Giannakeas, Nikolaos Tsipouras, Markos G. University of Western Macedonia Department of Electrical and Computer Engineering Kozani Greece University of Ioannina Department of Informatics and Telecommunications Ioannina Greece

ISBN: (纸本)9798350303964

EEG analysis for attention and memory assessment is an important research topic that has gained a lot of attention in recent years. Virtual Reality (VR) is an innovative technology that provides the ability to design experiments suitable for research in this field. This study explores the feasibility of utilizing ensemble classification algorithms to analyze EEG signals obtained during VR stimulation for the purpose of evaluating memory encoding and attention orienting. The obtained results indicate that EEG analysis can accurately assess memory and attention levels in VR environment setting with a high degree of certainty (84.90% accuracy, and an average of 0.86 true positive rate, 0.88 positive predicted value, 0.89 area under the ROC curve, and 0.89 true negative rate). © 2023 IEEE.

关键词： Virtual reality

来源：评论

学校读者我要写书评

暂无评论

Apply Combined Model Between VGG16 and LightGBM on Weather Image Recognition

Apply Combined Model Between VGG16 and LightGBM on Weather I...

引用

2023 RIVF International Conference on Computing and Communication Technologies, RIVF 2023

作者： Quy Nam, Tran Syamsuddin, Irfan Posts and Telecommunications Institute of Technology Hanoi Viet Nam Politeknik Negeri Ujung Department of Informatics & Computer Engineering Pandang Indonesia

ISBN: (纸本)9798350315844

This research implements hybrid deep learning network models for weather image classification. The study proposes to apply a combined model, namely VGG16-LightGBM. In its architecture model, a pre-train convolutional neural network (CNN) name as VGG16 is employed for feature extraction of images and the LightGBM algorithm is used to make classification. The results on accuracy of proposed models were compared with four other models, namely Xception, Inception V3,Vgg19, Vgg16 which are all implemented by transfer learning mechanism on the same dataset. The experimental results proved that the VGG16-LightGBM gives the best performance with the highest accuracy of 81,28%, outperforms the transfer learning technique of other 4 pre-train models in the problem of weather image classification. © 2023 IEEE.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Cluster-Based Method for Eavesdropping Identification and Localization in Optical Links

Cluster-Based Method for Eavesdropping Identification and Lo...

引用

2023 Asia Communications and Photonics Conference/2023 International Photonics and Optoelectronics Meetings, ACP/POEM 2023

作者： Song, Haokun Lin, Rui Sgambelluri, Andrea Cugini, Filippo Li, Yajie Zhang, Jie Monti, Paolo Chalmers University of Technology Electrical Engineering Department Gothenburg Sweden School of Electronic Engineering Beijing University of Posts and Telecommunications Gothenburg Sweden Institute of Telecommunications Informatics and Photonics Scuola Superiore Sant'Anna Pisa Italy

ISBN: (纸本)9798350312614

We propose a cluster-based method to detect and locate eavesdropping events in optical line systems characterized by small power losses. Our findings indicate that detecting such subtle losses from eavesdropping can be accomplished solely through optical performance monitoring (OPM) data collected at the receiver. On the other hand, the localization of such events can be effectively achieved by leveraging in-line OPM data. © 2023 IEEE.

关键词： eavesdropping machine learning power

来源：评论

学校读者我要写书评

暂无评论

Development of Gamification and Live Coding-Based Programming Learning Platform to Foster Learning Motivation of Vocational Students

Development of Gamification and Live Coding-Based Programmin...

引用

2023 International Conference on Electrical and Information Technology, IEIT 2023

作者： Hidayat, Wahyu Nur Dwi Prasetyo, Wiji Zahiroh, Ayu Najmatus Maydi Syahri, Esa Pramudheva Afdloly, Mochammad Mu'iz Sari, Rahajeng Kartika Universitas Negeri Malang Department of Electrical Engineering and Informatics Malang Indonesia Smk Negeri 10 Malang Departement of Computer Network Engineering and Telecommunications Malang Indonesia

ISBN: (纸本)9798350327298

The current utilization of e-learning has not been able to increase participation and learning experience and motivate students to achieve educational goals. This research focuses on developing gamification-based learning media in Object-Oriented Programming (OOP) subjects to increase student learning motivation, determine the feasibility level of learning media, and determine the increase in student learning motivation. The method of the research is design thinking It is used to identify and better understand users in the use of applications, find problems faced by users, and can explore new ideas in the creation of appropriate application features in accordance with user needs. The results of expert media validation from the teacher, lecturer, and programmer amounted to 89.78% in the good category, for the material validation results of 91.33% in the very good category. Media use was implemented by 23 students of class XI RPL SMKN 12 Malang. The user trial results showed a percentage of 83.48% with good criteria. The student learning motivation test results showed a percentage of 81.26% with very high criteria. This shows the effect of gamification implementation in fostering student learning motivation. Gamification-based media has been tested for feasibility by media experts, material experts, and students. Therefore, the media is feasible and can be used as a learning support media in object-oriented programming subjects in Vocational High School. © 2023 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Reconfiguring wireless environments via intelligent surfaces for 6G: reflection, modulation, and security

引用

Science China(Information Sciences) 2023年第3期66卷 60-79页

作者： Jindan XU Chau YUEN Chongwen HUANG Naveed UL HASSAN George C.ALEXANDROPOULOS Marco DI RENZO Mérouane DEBBAH Engineering Product Development Pillar Singapore University of Technology and Design College of Information Science and Electronic Engineering Zhejiang University Department of Electrical Engineering Lahore University of Management Sciences Department of Informatics and Telecommunications National and Kapodistrian University of Athens Technology Innovation Institute Université Paris-Saclay CNRS Centrale Supélec Laboratoire des Signaux et Systèmes

The reconfigurable intelligent surface(RIS) has been recognized as an essential enabling technology for sixth-generation(6G) mobile communication networks. An RIS comprises a large number of small and low-cost reflecting elements whose parameters can be dynamically adjusted with a programmable *** of these elements can effectively reflect a phase-shifted version of the incident electromagnetic wave. By configuring the wave phases in real time, the propagation environment of the information-bearing signals can be dynamically manipulated to enhance communication reliability, boost transmission rate, expand cellular coverage, and strengthen communication security. In this study, we provide an overview on RIS-assisted wireless communications. Specifically, we elaborate on the state-of-the-art enabling techniques for the RIS technology as well as their corresponding substantial benefits from the perspectives of RIS reflection and RIS modulation. With these benefits, we envision the integration of RISs into emerging applications for6G. In addition, communication security is of unprecedented importance in future 6G networks with ubiquitous wireless services in multifarious verticals and areas. We highlight potential contributions of RISs to physical-layer security, in terms of secrecy rate and secrecy outage probability, exemplified by a typical case study from both theoretical and numerical aspects. Finally, we discuss challenges and opportunities on the deployment of RISs in practice to motivate future research.

关键词： reconfigurable intelligent surface physical-layer security RIS-assisted wireless communications phase modulation 6G

来源：评论

学校读者我要写书评

暂无评论

CNN-JEPA: Self-Supervised Pretraining Convolutional Neural Networks Using Joint Embedding Predictive Architecture 23

CNN-JEPA: Self-Supervised Pretraining Convolutional Neural N...

引用

23rd IEEE International Conference on Machine Learning and Applications, ICMLA 2024

ISBN: (纸本)9798350374889

Self-supervised learning (SSL) has become an im-portant approach in pretraining large neural networks, enabling unprecedented scaling of model and dataset sizes. While recent advances like I-JEPA have shown promising results for Vision Transformers, adapting such methods to Convolutional Neural Networks (CNNs) presents unique challenges. In this paper, we introduce CNN -JEPA, a novel SSL method that successfully applies the joint embedding predictive architecture approach to CNN s. Our method incorporates a sparse CNN encoder to handle masked inputs, a fully convolutional predictor using depthwise separable convolutions, and an improved masking strategy. We demonstrate that CNN-JEPA outperforms I-JEPA with ViT architectures on ImageNet-l00, achieving a 73.3% linear top-1 accuracy using a standard ResN et-50 encoder. Compared to other CNN-based SSL methods, CNN-JEPA requires 17- 35 % less training time for the same number of epochs and approaches the linear and k-NN top-l accuracies of BYOL, SimCLR, and VICReg. Our approach offers a simpler, more efficient alternative to existing SSL methods for CNNs, requiring minimal augmentations and no separate projector network. © 2024 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Road Traffic Safety Status Analysis and Prediction Based on Dynamic Bayesian Network

Road Traffic Safety Status Analysis and Prediction Based on ...

引用

2024 International Conference on Computational Intelligence and Communication System, CCICS 2024

作者： Zhao, Pengzhi Geng, Danyang She, Shaoyi Duan, Mengyuan China Transport Telecommunications & Information Center Beijing100011 China China Transport Informatics National Engineering Laboratory Co. Ltd. Beijing100094 China

Dynamic Bayesian networks can effectively capture dynamic changes and uncertainty relationships in data. Conventional prediction methods do not consider the temporal characteristics between traffic flow sequences, which affects prediction accuracy. This article proposes a method for analyzing and predicting road traffic safety status based on DBN. Firstly, data matching is performed according to the "case-control"sample structure of the matching formula to minimize the influence of other factors on the modeling of traffic safety status;Secondly, the random forest model is applied to analyze and extract the variable with the highest correlation coefficient as the input variable for the traffic safety status prediction model;Then, a DBN prediction model is established using matched accident traffic flow and non-accident traffic flow sample data;Finally, by analyzing the effectiveness evaluation indicators of the model, multiple prediction results showed that the overall prediction accuracy of the DBN method was over 80%. © Published under licence by IOP Publishing Ltd.

关键词： Decision trees

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：