检索结果-内蒙古大学图书馆

6th IEEE International Conference on Power, Intelligent Computing and Systems, ICPICS 2024

作者： Yu, Yichun Yang, Xiaoyi Lan, Yuqing Xing, Zhihuan Yu, Dan Zheng, Han Wang, Danxing School of Computer Science and Engineering Beihang University Beijing China School of Software Beihang University Beijing China China Standard Intelligent Security Beijing China

ISBN: (纸本)9798350374315

With the rapid proliferation of IoT (Internet of Things) technologies, ensuring reliable and continuous operation of M2M (machine to machine) communications has become paramount. This paper presents the design and implementation of a redundant backup system tailored for M2M devices, utilizing an ad hoc network system. The proposed architecture is systematically divided into three layers: the physical layer, network layer, and application layer, each meticulously crafted to address the specific constraints and demands inherent to IoT devices. At the physical layer, considerations are given to the hardware specifications and energy consumption of M2M devices. The network layer focuses on establishing robust, self-organizing ad hoc network protocols to ensure seamless connectivity and data transmission between devices. Meanwhile, the application layer is designed to manage data redundancy, recovery processes, and system monitoring, ensuring that the backup mechanism operates efficiently under varying conditions. Through the integration of this multi-layered device redundant backup solution, the system guarantees continuous operation and enhances reliability in M2M scenarios. Extensive testing and simulation results demonstrate the effectiveness of the proposed system in maintaining data integrity and operational resilience, even in the face of network disruptions and hardware failures. This study offers a comprehensive framework for improving the robustness of M2M communications, contributing to the advancement of IoT technologies. © 2024 IEEE.

关键词： Ad hoc networks

来源：评论

学校读者我要写书评

暂无评论

Effective Training of PINNs by Combining CMA-ES with Gradient Descent 13

Effective Training of PINNs by Combining CMA-ES with Gradien...

引用

13th IEEE Congress on Evolutionary Computation, CEC 2024

作者： Liu, Lin Yuan, Yuan School of Computer Science and Engineering Beihang University Beijing100191 China State Key Laboratory of Software Development Environment Zhongguancun Laboratory Beijing100191 China

ISBN: (纸本)9798350308365

Physics-Informed Neural Networks (PINNs) have recently received increasing attention, however, optimizing the loss function of PINNs is notoriously difficult, where the landscape of the loss function is often highly non-convex and rugged. Local optimization methods based on gradient information can converge quickly but are prone to being trapped in local minima for training PINNs. Evolutionary algorithms (EAs) are well known for the global search ability, which can help escape from local minima. It has been reported in the literature that EAs show some advantages over gradient-based methods in training PINNs. Inspired by the Memetic Algorithm, we combine global-search based EAs and local-search based batch gradient descent in order to make the best of both word. In addition, since the PINN loss function is composed of multiple terms, balancing these terms is also a challenging issue. Therefore, we also attempt to combine EAs with multiple-gradient descent algorithm for multi-objective optimization. Our experiments provide strong evidence for the superiority of the above algorithms. © 2024 IEEE.

关键词： Optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Plausible Heterogeneous Graph k-Anonymization for Social Networks

引用

Tsinghua science and Technology 2022年第6期27卷 912-924页

作者： Kaiyang Li Ling Tian Xu Zheng Bei Hui School of Computer Science and Engineering University of Electronic Science and Technology of ChinaChengdu 611731China School of Information and Software Engineering University of Electronic Science and Technology of ChinaChengdu 610054China Trusted Cloud Computing and Big Data Key Laboratory of Sichuan Province Chengdu 610000China

The inefficient utilization of ubiquitous graph data with combinatorial structures necessitates graph embedding methods,aiming at learning a continuous vector space for the graph which is amenable to be adopted in traditional machine learning algorithms in favor of vector *** embedding methods build an important bridge between social network analysis and data analytics as social networks naturally generate an unprecedented volume of graph data *** social network data not only bring benefit for public health,disaster response,commercial promotion,and many other applications,but also give birth to threats that jeopardize each individual’s privacy and ***,most existing works in publishing social graph embedding data only focus on preserving social graph structure with less attention paid to the privacy issues inherited from social *** be specific,attackers can infer the presence of a sensitive relationship between two individuals by training a predictive model with the exposed social network *** this paper,we propose a novel link-privacy preserved graph embedding framework using adversarial learning,which can reduce adversary’s prediction accuracy on sensitive links while persevering sufficient non-sensitive information such as graph topology and node attributes in graph *** experiments are conducted to evaluate the proposed framework using ground truth social network datasets.

关键词： social network graph embedding privacy preservation adversarial learning

来源：评论

学校读者我要写书评

暂无评论

Insider Attack Detection Using Deep Belief Neural Network in Cloud Computing

引用

computer Systems science & engineering 2022年第5期41卷 479-492页

作者： A.S.Anakath R.Kannadasan Niju P.Joseph P.Boominathan G.R.Sreekanth School of Computing E.G.S.Pillay Engineering CollegeNagapattinam611002TamilnaduIndia Department of Software Systems School of Computer Science and Engineering(SCOPE)Vellore Institute of Technology UniversityVellore632014TamilnaduIndia Department of Computer Science CHRIST(Deemed to be University)Bengaluru560029India Department of Information Security School of Computer Science and Engineering(SCOPE)Vellore Institute of Technology UniversityVellore632014TamilnaduIndia Department of Computer Science and Engineering Kongu Engineering CollegeErode638060TamilnaduIndia

Cloud computing is a high network infrastructure where users,owners,third users,authorized users,and customers can access and store their information *** use of cloud computing has realized the rapid increase of information in every field and the need for a centralized location for processing *** cloud is nowadays highly affected by internal threats of the *** applications such as banking,hospital,and business are more likely affected by real user *** intruder is presented as a user and set as a member of the *** becoming an insider in the network,they will try to attack or steal sensitive data during information sharing or *** major issue in today's technological development is identifying the insider threat in the cloud *** data are lost,compromising cloud users is *** and security are not ensured,and then,the usage of the cloud is not *** solutions are available for the external security of the cloud ***,insider or internal threats need to be *** this research work,we focus on a solution for identifying an insider attack using the artificial intelligence *** insider attack is possible by using nodes of weak users’*** will log in using a weak user id,connect to a network,and pretend to be a trusted ***,they can easily attack and hack information as an insider,and identifying them is very *** types of attacks need intelligent solutions.A machine learning approach is widely used for security *** date,the existing lags can classify the attackers *** information hijacking process is very absurd,which motivates young researchers to provide a solution for internal *** our proposed work,we track the attackers using a user interaction behavior pattern and deep learning *** usage of mouse movements and clicks and keystrokes of the real user is stored in a *** deep belief neural

关键词： Cloud computing security insider attack network security privacy user interaction behavior deep belief neural network

来源：评论

学校读者我要写书评

暂无评论

An Investigation into Improved YOLOv8-based Target Detection Algorithms for UAV Aerial Imagery 2

An Investigation into Improved YOLOv8-based Target Detection...

引用

2nd International Conference on Image Processing, computer Vision and Machine Learning, ICICML 2023

作者： Ming, Rui Liu, Xinyu Lin, Yuxuan Gong, Qian Minjiang University College of Computer and Control Engineering Fuzhou350108 China Fuzhou University College of Computer and Data Science College of Software Fuzhou350108 China

ISBN: (纸本)9798350331417

In UAV imagery, the intricate backgrounds combined with the high quantity and compact distribution of minute targets have consistently made target detection a formidable challenge in the realm of computer vision. This study introduces an enhancement over the YOLOv8 algorithm, wherein a sophisticated multi-scale convolutional layer, integrating depth-separable convolution, attention mechanisms, and multi-scale processing techniques, replaces the original model's convolution. Moreover, we introduce an attention mechanism for a Bi-Level Routing within the core component of the base model, and adjustments are made to the original model's loss function. Lastly, to confirm the viability of the enhanced model proposed in this paper, we conducted a validation of the metrics using publicly accessible datasets. The findings illustrate that the improved model outlined in this research substantially enhances target recognition accuracy in UAV images. Furthermore, the model exhibits superior performance in mitigating issues of duplicate detection and target omission. © 2023 IEEE.

关键词： Multi-scale Convolutions Target Detection UAV Aerial Imagery YOLOv8-based

来源：评论

学校读者我要写书评

暂无评论

PREDICTION UNCERTAINTY FLATTENING FOR PERFORMANCE IMPROVEMENT OF DEEP LEARNING 5

PREDICTION UNCERTAINTY FLATTENING FOR PERFORMANCE IMPROVEMEN...

引用

5th ECCOMAS Thematic Conference on Uncertainty Quantification in Computational sciences and engineering, UNCECOMP 2023

作者： Liu, Chao Zhou, Xinlei Wang, Xizhao College of Computer Science and Software Engineering Shenzhen University 3688 Nanhai Avenue Nanshan District Guangdong Province Shenzhen China

The representation, measure, and handling of uncertainty in an application of deep learning have a significant impact on the performance of learning systems. Uncertainty is embedded in the entire process of learning from data, which can roughly be categorized as three types, i.e., data uncertainty, model uncertainty, and prediction uncertainty. Appropriately modeling and processing uncertainties in different phases of learning can significantly improve accuracy and robustness. Focusing on the 3rd type of uncertainty, i.e., the prediction uncertainty which lies on the phase from the model output (i.e., a probability distribution on label space) to the final class(s) determination, we present in this paper a new viewpoint regarding uncertainty processing. Theoretically and experimentally, it establishes a statistical relationship between the prediction uncertainty and the learning performance, which implicitly points out that some results about uncertainty processing in traditional textbooks may be incorrect. Based on this viewpoint, we propose an approach to optimize the output layer of neural network models. The method improves the prediction accuracy by replacing the last layer of softmax with a random weight model which is a simple and fast non-iterative training algorithm. Extensive experiments and simulations on the synthetic and real datasets demonstrate the effectiveness and good generalization of the proposed scheme. © 2023 UNCECOMP Proceedings. All rights reserved.

关键词： Probability distributions

来源：评论

学校读者我要写书评

暂无评论

Automated computer Vision-based Detection of Solar Panel Defects Using a Thermal Camera Mobile Application 10

Automated Computer Vision-based Detection of Solar Panel Def...

引用

10th International Conference on Future Internet of Things and Cloud, FiCloud 2023

作者： Hajjdiab, Hassan Mohamed, Gasm ELbary Alzaabi, Salim Alhalabi, Marah Gad, Abdalla Yaghi, Maha Alkhedher, Mohammad Ghazal, Mohammed Concordia University Department of Computer Science and Software Engineering Montreal Canada Abu Dhabi University College of Engineering Abu Dhabi United Arab Emirates

ISBN: (纸本)9798350316353

Solar energy production has risen significantly globally in recent years. Solar Panels and Photovoltaic PV modules are the key components of solar energy production. Therefore, ensuring their health and efficiency and finding technologies to facilitate this became more important. However, the PV inspection costs money, time, and human resources. Thus, many technologies and techniques have been introduced to facilitate PV cell inspection, such as using a drone with a thermal camera. Moreover, as technology evolves every day, so do mobile devices. Nowadays, smartphones have brought many conveniences to our lives. Smartphones have become smarter with applications, and now they can be used as health monitoring devices, projectors, and more. One of smartphones' most revolutionary features and technology is that they can come with a thermal camera or be compatible with thermal cameras. Thermal camera application varies and is diverse, from detecting heat on an object to inspecting photovoltaic PV cells. This paper proposes a novel system consisting of a thermal camera mobile app to detect the defects in PV modules and estimate the defect percentage. The result of this work has shown a potential research and further development area in the field of renewable energy and PV cell inspection. © 2023 IEEE.

关键词： Solar panels

来源：评论

学校读者我要写书评

暂无评论

A Survey on Metaverse: Applications, Crimes and Governance

A Survey on Metaverse: Applications, Crimes and Governance

引用

2023 IEEE International Conference on Metaverse Computing, Networking and Applications, MetaCom 2023

作者： Lin, Kaixin Wu, Jiajing Lin, Dan Zheng, Zibin Sun Yat-Sen University School of Computer Science and Engineering Guangzhou China Sun Yat-Sen University School of Software Engineering Zhuhai China

ISBN: (纸本)9798350333336

With the rapid development of blockchain and the increasing computing power, the realization of the metaverse has become possible. However, there are not yet well-regulated and standardized industry standards for the metaverse, thus various financial crimes have been bred in many metaverse projects: scams, code exploits, wash trading, money laundering, illegal services, shops, etc. For this reason, it is urgent to categorize the crimes that exist in the metaverse ecosystem and summarize the corresponding governance. In this paper, we first describe the background and applications of the metaverse. Next, we present detailed case studies on different types of crimes in the metaverse, and summarize the metaverse's current regulatory measures, related governance. Finally, we suggest possible directions and recommendations for the development of relevant measures and governance. The overview of this paper is expected to help readers gain a deeper understanding of the financial threats in the metaverse. © 2023 IEEE.

关键词： Crime

来源：评论

学校读者我要写书评

暂无评论

Neural Pathways to Attention Enhancement: EEG Spectral Ratio Analysis of CM-II Meditation's Effect on Student Attention 54

Neural Pathways to Attention Enhancement: EEG Spectral Ratio...

引用

54th IEEE Frontiers in Education Conference, FIE 2024

作者： Gopi, Sreekanth Dehbozorgi, Nasrin Kennesaw State University Department of Computer Science KennesawGA United States Kennesaw State University Department of Software Engineering KennesawGA United States

ISBN: (纸本)9798350351507

This research-to-practice paper describes the development of an innovative practice aimed at addressing the global surge in student stress and related mental health challenges, particularly in engineering education. Student well-being is foundational to academic success, with studies indicating that dropout rates for students with stress are more than 50% and mental health problems can range as high as 83% in engineering education. Stress and elevated levels of inattention are associated with decreased academic interest, decision-making, self-efficacy, study skills, and GPA. Research has shown that stress levels can be reduced by meditation techniques and they induce attention. In this research, a three-stage guided meditation called ChakraMarmaKosha Meditation II (CM-II) was developed to improve the attention level of students. It involves past emotional cleansing, reviewing the day, visualizing the future components, and is accompanied by flute music that is composed based on a chosen Raga. In this study, a data analysis pipeline was developed for detailed spectral analysis to measure variations in attention due to meditation. In an in-person EEG experiment, a study group of 15 college students underwent this CM-II meditation process in a lab setting. The EEG spectral analysis underscored an enhancement in attentional focus post-meditation, as evidenced by key shifts in spectral ratios such as the notable decrease in Theta to Beta Ratio (TBR) and the surge in Theta to Gamma Ratio (TGR). The observed results were validated using attention scores from tests on the TestMyBrain tool. This study underscores the potential benefits of the CM-II meditation technique in fostering enhanced attention, reduced stress, and consequently improved academic performance in engineering education. While the current study provides promising results with CM-II meditation in enhancing attention, cognition, and relaxation, it's essential to acknowledge the limitations, such as the confin

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

DocTTT: Test-Time Training for Handwritten Document Recognition Using Meta-Auxiliary Learning

arXiv

引用

arXiv 2025年

作者： Gu, Wenhao Gu, Li Wang, Ziqiang Suen, Ching Yee Wang, Yang Department of Computer Science and Software Engineering Concordia University Canada

Despite recent significant advancements in Handwritten Document Recognition (HDR), the efficient and accurate recognition of text against complex backgrounds, diverse handwriting styles, and varying document layouts remains a practical challenge. Moreover, this issue is seldom addressed in academic research, particularly in scenarios with minimal annotated data available. In this paper, we introduce the DocTTT framework to address these challenges. The key innovation of our approach is that it uses test-time training to adapt the model to each specific input during testing. We propose a novel Meta-Auxiliary learning approach that combines Meta-learning and self-supervised Masked Autoencoder (MAE). During testing, we adapt the visual representation parameters using a self-supervised MAE loss. During training, we learn the model parameters using a meta-learning framework, so that the model parameters are learned to adapt to a new input effectively. Experimental results show that our proposed method significantly outperforms existing state-of-the-art approaches on benchmark datasets. Copyright © 2025, The Authors. All rights reserved.

关键词： Self-supervised learning

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：