检索结果-内蒙古大学图书馆

Proceedings of the 41st International Conference on Machine Learning

作者： Cong Hua Qianqian Xu Shilong Bao Zhiyong Yang Qingming Huang Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China and School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China Institute of Information Engineering Chinese Academy of Sciences Beijing China and School of Cyber Security University of Chinese Academy of Sciences Beijing China School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China and Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China and Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing China

This paper explores a novel multi-modal alternating learning paradigm pursuing a reconciliation between the exploitation of uni-modal features and the exploration of cross-modal interactions. This is motivated by the fact that current paradigms of multi-modal learning tend to explore multi-modal features simultaneously. The resulting gradient prohibits further exploitation of the features in the weak modality, leading to modality competition, where the dominant modality overpowers the learning process. To address this issue, we study the modality-alternating learning paradigm to achieve reconcilement. Specifically, we propose a new method called ReconBoost to update a fixed modality each time. Herein, the learning objective is dynamically adjusted with a reconcilement regularization against competition with the historical models. By choosing a KL-based reconcilement, we show that the proposed method resembles Friedman's Gradient-Boosting (GB) algorithm, where the updated learner can correct errors made by others and help enhance the overall performance. The major difference with the classic GB is that we only preserve the newest model for each modality to avoid overfitting caused by ensembling strong learners. Furthermore, we propose a memory consolidation scheme and a global rectification scheme to make this strategy more effective. Experiments over six multi-modal benchmarks speak to the efficacy of the method. We release the code at https://***/huacong/ReconBoost.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Shield-U: Safeguarding Traffic Sign Recognition Against Perturbation Attacks

Shield-U: Safeguarding Traffic Sign Recognition Against Pert...

引用

IEEE International Conference on Trust, Security and Privacy in computing and Communications (TrustCom)

作者： Shengmin Xu Jianfei Sun Hangcheng Cao Yulan Gao Ziyang He Cong Wu College of Computer and Cyber Security Fujian Normal University China School of Computing and Information Systems Singapore Management University Singapore Department of Computer Science City University of Hong Kong China Division of Information Science and Engineering KTH Royal Institute of Technology Sweden School of Cyber Science and Engineering Zhengzhou University China School of Computer Science and Engineering Nanyang Technological University Singapore

ISBN: (数字)9798331506209

ISBN: (纸本)9798331506216

Traffic sign recognition systems are crucial for the navigation and situation awareness of autonomous vehicles. They leverage deep learning technologies to swiftly and accurately identify traffic signs, even in the most challenging traffic environments. However, security researchers have uncovered a critical vulnerability in these systems: learning-based TSRs are particularly susceptible to physical-world perturbation attacks. Through subtle modifications (i.e., attaching well-designed patches on traffic signs), attackers can deceive the recognition system into making erroneous judgments, which can further lead to serious traffic accidents. Although several defense mechanisms have been proposed to enhance the security of sign recognition systems, these solutions generally target only specific types of malicious perturbations and thus lack robustness. To address this issue, we present a robust defense mechanism named Shield-U, which restores traffic sign images contaminated by physical patch perturbations, providing credible data for the recognition model. In the process of implementing Shield-U, we first design a feature difference-aware perturbation generator that outputs potential sign contamination patterns. Incorporating generated perturbations during the training phase enables our restoration model to gain sufficient understanding of diverse perturbation types, thus enhancing its ability to repair various perturbed signs. Following this, we build an attention-driven restoration network to repair sign images. Finally, we evaluate the effectiveness of Shield-U using widely used sign recognition models and public datasets. The results demonstrate that our defense mechanism excels in resisting potential perturbations, increasing the average sign recognition accuracy by 50.4%.

关键词： Training Image recognition Target recognition Perturbation methods Maintenance engineering Robustness Image restoration Safety Security Public transportation

来源：评论

学校读者我要写书评

暂无评论

Wafer Surface Semiconductor Defect Classification Using Convolution Neural Network Based Improved Faster R-CNN

Wafer Surface Semiconductor Defect Classification Using Conv...

引用

Data science and Information System (ICDSIS), IEEE International Conference on

作者： Praveenkumar Babu Mohammed I. Habelalmateen Ramya Srikanteswara R. Archana Reddy N. Purushotham Department of Electronics and Communication Engineering SRM Institute of Science and Technology Ramapuram Chennai India Department of Computers Techniques Engineering College of Technical Engineering The Islamic University Najaf Iraq Department of Computer Science and Engineering Nitte Meenakshi Institute of Technology Bengaluru India School of Computer Science and Engineering SR University Warangal Telangana India Dept. of CSE School of Computing Mohan Babu University Tirupati India

ISBN: (数字)9798350381665

ISBN: (纸本)9798350381672

Defect classification helps ensure the quality and reliability of semiconductor devices. Even a minor defect can lead to significant performance issues or failures in electronic components. Moreover, a recurring issue in this field is the identification of minute fault patterns, frequently measuring just a few $\mu \mathrm{m}$ in size, and pixels among enormous volumes of high-resolution imaging. In this paper, proposed the improved Faster Region-Based Convolutional Neural Network (R-CNN) by utilizing the Res2Net model can assign the backbone of the proposed model. The model can train and validating on WM811 K dataset. The dataset contains imbalance classes in the presented categories. Initially the data-augmentation technique is employed to create the synthetic data, and it solve the classes imbalance issue. Subsequently Build the Convolutional Neural Network (CNN) foe classify the categories of the defects by using the extracted features. The Res2Net is effectively extract the various features and the feature pyramid network is used to fuse the extracted information. The proposed method is evaluated and compared with various existing methods and metrices such as Accuracy of $97.86 \%$, precision of $96.50 \%$, Recall of $96.80 \%$ and $F 1$-score of $97.85 \%$ and the methods are namely Deep Convolutional Neural Network for Wafer Defect Identification (DCNN-WDI), ShuffleNet-v2-CNN and Transfer Learning Using Faster R-CNN.

关键词： Semiconductor device modeling Accuracy Volume measurement Transfer learning Feature extraction Data models Numerical models

来源：评论

学校读者我要写书评

暂无评论

Software Complexity Reduction through the Process Automation in Software Development Life Cycle 4

Software Complexity Reduction through the Process Automation...

引用

4th International Conference on Electrical, computer and Communication Technologies, ICECCT 2021

作者： Wijendra, Dinuka R. Hewagamage, K.P. Sri Lanka Institute of Information Technology Department of Information Technology Malabe Sri Lanka University of Colombo School of Computing Computer Science Colombo Sri Lanka

ISBN: (纸本)9781665414807

Numerous software complexity metrics have been introduced to quantify the software complexity in terms of different attributes considered in its written source code. Although the complexity determination is bounded with its source code, it should be expressed beyond its code base level, since the software is implemented as a combination of different phrases inside the Software Developments Life Cycle. The automation of the processes involved in software implementation procedure will mitigate the human effort taken during the phrases, resulting that the overall complexity of the software will also be reduced. The proposed system has the capability to demonstrate the requirement analysis, design, defects tracking, quality analysis and the complexity computation with respective to the different complexity metrics without restraining the software complexity evaluation into several quality attributes within the source code. © 2021 IEEE.

关键词： Quality control

来源：评论

学校读者我要写书评

暂无评论

ExplainableDetector: Exploring Transformer-based Language Modeling Approach for SMS Spam Detection with Explainability Analysis

arXiv

引用

arXiv 2024年

作者： Uddin, Mohammad Amaz Islam, Muhammad Nazrul Maglaras, Leandros Janicke, Helge Sarker, Iqbal H. Department of Computer Science and Engineering BGC Trust University Bangladesh Chittagong4381 Bangladesh Department of Computer Science and Engineering Military Institute of Science and Technology Dhaka1216 Bangladesh School of Computing Edinburgh Napier University United Kingdom Centre for Securing Digital Futures Edith Cowan University PerthWA6027 Australia

SMS, or short messaging service, is a widely used and cost-effective communication medium that has sadly turned into a haven for unwanted messages, commonly known as SMS spam. With the rapid adoption of smartphones and Internet connectivity, SMS spam has emerged as a prevalent threat. Spammers have taken notice of the significance of SMS for mobile phone users. Consequently, with the emergence of new cybersecurity threats, the number of SMS spam has expanded significantly in recent years. The unstructured format of SMS data creates significant challenges for SMS spam detection, making it more difficult to successfully fight spam attacks in the cybersecurity domain. In this work, we employ optimized and fine-tuned transformer-based Large Language Models (LLMs) to solve the problem of spam message detection. We use a benchmark SMS spam dataset for this spam detection and utilize several preprocessing techniques to get clean and noise-free data and solve the class imbalance problem using the text augmentation technique. The overall experiment showed that our optimized fine-tuned BERT (Bidirectional Encoder Representations from Transformers) variant model RoBERTa obtained high accuracy with 99.84%. We also work with Explainable Artificial Intelligence (XAI) techniques to calculate the positive and negative coefficient scores which explore and explain the fine-tuned model transparency in this text-based spam SMS detection task. In addition, traditional Machine Learning (ML) models were also examined to compare their performance with the transformer-based models. This analysis describes how LLMs can make a good impact on complex textual-based spam data in the cybersecurity field. © 2024, CC BY.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Dual-scale Enhanced and Cross-generative Consistency Learning for Semi-supervised Medical Image Segmentation

arXiv

引用

arXiv 2023年

作者： Gu, Yunqi Zhou, Tao Zhang, Yizhe Zhou, Yi He, Kelei Gong, Chen Fu, Huazhu PCA Laboratory The School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing210094 China The School of Computer Science and Engineering Southeast University Nanjing211189 China The Medical School Nanjing University Nanjing210023 China The National Institute of Healthcare Data Science Nanjing University Nanjing210023 China The Institute of High Performance Computing A*STAR Singapore

Medical image segmentation plays a crucial role in computer-aided diagnosis. However, existing methods heavily rely on fully supervised training, which requires a large amount of labeled data with time-consuming pixel-wise annotations. Moreover, accurately segmenting lesions poses challenges due to variations in shape, size, and location. To address these issues, we propose a novel Dual-scale Enhanced and Cross-generative consistency learning framework for semi-supervised medical image Segmentation (DEC-Seg). First, we propose a Cross-level Feature Aggregation (CFA) module that integrates cross-level adjacent layers to enhance the feature representation ability across different resolutions. To address scale variation, we present a scale-enhanced consistency constraint, which ensures consistency in the segmentation maps generated from the same input image at different scales. This constraint helps handle variations in lesion sizes and improves the robustness of the model. Furthermore, we propose a cross-generative consistency scheme, in which the original and perturbed images can be reconstructed using cross-segmentation maps. This consistency constraint allows us to mine effective feature representations and boost the segmentation performance. To further exploit the scale information, we propose a Dual-scale Complementary Fusion (DCF) module that integrates features from two scale-specific decoders operating at different scales to help produce more accurate segmentation maps. Extensive experimental results on multiple medical segmentation tasks (polyp, skin lesion, and brain glioma) demonstrate the effectiveness of our DECSeg against other state-of-the-art semi-supervised segmentation approaches. The implementation code will be released at https://***/taozh2017/DECSeg. Copyright © 2023, 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

Adaptive Power Control Aware Depth Routing in Underwater Sensor Networks

引用

computers, Materials & Continua 2021年第10期69卷 1301-1322页

作者： Ghufran Ahmed Saiful Islam Ihsan Ali Isra Adil Hayder Abdelmuttlib Ibrahim Abdalla Ahmed Muhammad Talha Sultan S.Alshamrani Ag Asri Ag Ibrahim Department of Computer Science FAST National University of Computer and Emerging Sciences(NUCES)KarachiPakistan Dr.A.Q.Khan Institute of Computer Sciences and Information Technology KahutaPakistan Faculty of Computer Science and Information Technology University of MalayaKuala Lumpur50603Malaysia Department of Computer Science COMSATS University IslamabadIslamabadPakistan Deanship of Scientific Research King Saud UniversityRiyadhSaudi Arabia Department of Information Technology College of Computer and Information TechnologyTaif UniversityTaif21944Saudi Arabia Faculty of Computing and Informatics Universiti Malaysia SabahKota KinabaluSabahMalaysia

Underwater acoustic sensor network(UASN)refers to a procedure that promotes a broad spectrum of aquatic *** can be practically applied in seismic checking,ocean mine identification,resource exploration,pollution checking,and disaster *** confronts many difficulties and issues,such as low bandwidth,node movements,propagation delay,3D arrangement,energy limitation,and high-cost production and arrangement costs caused by antagonistic underwater *** wireless sensor networks(UWSNs)are considered a major issue being encountered in energy management because of the limited battery power of their ***,the harsh underwater environment requires vendors to design and deploy energy-hungry devices to fulfil the communication requirements and maintain an acceptable quality of ***,increased transmission power levels result in higher channel interference,thereby increasing packet *** the facts mentioned above,this research presents a controlled transmission power-based sparsity-aware energy-efficient clustering in *** contributions of this technique is ***,it uses the adaptive power control mechanism to utilize the sensor nodes’battery and reduce channel interference ***,thresholds are defined to ensure successful ***,clustering can be implemented in dense areas to decrease the repetitive transmission that ultimately affects the energy consumption of nodes and interference ***,mobile sinks are deployed to gather information locally to achieve the previously mentioned *** suggested protocol is meticulously examined through extensive simulations and is validated through comparison with other advanced UWSN *** show that the suggested protocol outperforms other procedures in terms of network lifetime and packet delivery ratio.

关键词： UWSNs UASNs QoS

来源：评论

学校读者我要写书评

暂无评论

Detection and Counter Measure of Packet Misrouting by GPS-Free Positioning of Ad-Hoc Nodes

引用

Mobile Information Systems 2023年第1期2023卷

作者： Sathianesan, Godfrey Winster Gnanavel, S. Salini, R. Raji, V. Vijay Anand, R. Uddin, Md Salah Department of Computing Technologies School of Computing SRM Institute of Science and Technology Tamil Nadu Kattankulathur603203 India Department of Computer Science and Engineering Panimalar Engineering College Tamil Nadu Chennai India Department of Computer Science and Engineering S K P Engineering College Tamil Nadu Tiruvannamalai606611 India School of Information Technology and Engineering Vellore Institute of Technology Tamil Nadu Vellore India Department of Multimedia and Creative Technology Daffodil International University Dhaka1207 Bangladesh

The ad-hoc networks are a developing area of research with a large and wide variety of applications and related requirements. The nodes participating in an ad-hoc network use different routing protocols to send their packets from one node to another. But most of the time, the ad-hoc network is not suitable for urgent needs. For this, an ad-hoc network GPS-free positioning system can be used in emergency situations to save people in danger. Nodes participating in an ad-hoc network choose the best route from various nodes near them to send information through the complex system. For this, we have used trust dynamic source routing (TDSR) in our research work to determine and design the best route to transmit information, and we designed this system with the aim of being able to be used in emergencies. We have compared our proposed routing protocols with dynamic source routing (DSR) and found out that TDSR is working excellent. © 2023 Godfrey Winster Sathianesan et al.

关键词： Ad hoc networks

来源：评论

学校读者我要写书评

暂无评论

A Critical Analysis of Object Tracking Using Radial Function Networks

A Critical Analysis of Object Tracking Using Radial Function...

引用

Smart Devices (ICSD), International Conference on

作者： Amit Kumar Mishra Neeraj Kumar Pandey Manoj Diwakar Nilesh Bhaskarrao Bahadure Ankur Dumka Jagdish Chandra Patni Dept. of Computer Science & Engineering Graphic Era Hill University Dehradun India Dept. of Computer Science & Engineering Graphic Era Deemed to be University Dehradun India Dept. of CSE Symbiosis Institute of Technology Nagpur Campus Symbiosis International (Deemed University) Pune India Dept. of Computer Science & Engineering Women Institute of Technology Dehradun India Dept. of CSE Alliance School of Advance Computing Alliance University Bengaluru Karnataka India

ISBN: (数字)9798350394412

ISBN: (纸本)9798350394429

A important method in several fields, including driverless cars, medical imaging, human-computer interaction, and investigation systems, is object tracking. Because of their flexibility to changes in object presence and ambient factors, as well as their capacity to estimated complex, non-linear functions, radial networks have become a potent tool for object stalking in recent years. The article conditions for the analysis of the radical working networking where the algorithm is a tracking segment that aims in moving the object. Object recognition is one of the most basic and centralized factor analyses in determining outputs having relevant values and a strong research background. Radial functioning is a phenomenon that has gained rapid training and desirable properties for improving the presentation of networking sites. radial networks, a form of artificial neural network, utilize radial basis functions as initiation functions. Because of this characteristic, RBF networks are especially well-suited for applications like object tracking that call for high accuracy in function approximation and pattern recognition. We explore the design of RBF networks in detail, emphasizing the functions of the input layer, output layer, and hidden layer using radial basis functions. We examine the mathematical foundations that allow RBF networks to represent intricate connections in the data, emphasizing how well they can generalize and interpolate.

关键词： Training Accuracy Supervised learning Radial basis function networks Real-time systems Pattern recognition Object tracking Object recognition Smart devices Logistics

来源：评论

学校读者我要写书评

暂无评论

NeRF-Texture: Synthesizing Neural Radiance Field Textures

arXiv

引用

arXiv 2024年

作者： Huang, Yi-Hua Cao, Yan-Pei Lai, Yu-Kun Shan, Ying Gao, Lin Beijing Key Laboratory of Mobile Computing and Pervasive Device Institute of Computing Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences Beijing China VAST ARC Lab Tencent PCG China School of Computer Science & Informatics Cardiff University United Kingdom

Texture synthesis is a fundamental problem in computer graphics that would benefit various applications. Existing methods are effective in handling 2D image textures. In contrast, many real-world textures contain meso-structure in the 3D geometry space, such as grass, leaves, and fabrics, which cannot be effectively modeled using only 2D image textures. We propose a novel texture synthesis method with Neural Radiance Fields (NeRF) to capture and synthesize textures from given multi-view images. In the proposed NeRF texture representation, a scene with fine geometric details is disentangled into the meso-structure textures and the underlying base shape. This allows textures with meso-structure to be effectively learned as latent features situated on the base shape, which are fed into a NeRF decoder trained simultaneously to represent the rich view-dependent appearance. Using this implicit representation, we can synthesize NeRF-based textures through patch matching of latent features. However, inconsistencies between the metrics of the reconstructed content space and the latent feature space may compromise the synthesis quality. To enhance matching performance, we further regularize the distribution of latent features by incorporating a clustering constraint. In addition to generating NeRF textures over a planar domain, our method can also synthesize NeRF textures over curved surfaces, which are practically useful. Experimental results and evaluations demonstrate the effectiveness of our approach. Copyright © 2024, The Authors. All rights reserved.

关键词： Image texture

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：