检索结果-内蒙古大学图书馆

38th Conference on Neural information Processing systems, NeurIPS 2024

作者： Wang, Jingjing Huang, Minhuan Nie, Yuanpin Li, Xiang Du, Qianjin Kong, Wei Deng, Huan Kuang, Xiaohui Institute of Systems Engineering Academy of Military Sciences PLA China Department of Computer Science and Technology Tsinghua University China School of Information Science and Engineering Zhejiang Sci-Tech University China

Deep learning technologies have demonstrated remarkable performance in vulnerability detection. Existing works primarily adopt a uniform and consistent feature learning pattern across the entire target set. While designed for general-purpose detection tasks, they lack sensitivity towards target code comprising multiple functional modules or diverse vulnerability subtypes. In this paper, we present a knowledge fusion-based vulnerability detection method (KF-GVD) that integrates specific vulnerability knowledge into the Graph Neural Network feature learning process. KF-GVD achieves accurate vulnerability detection across different functional modules of the Linux kernel and vulnerability subtypes without compromising general task performance. Extensive experiments demonstrate that KF-GVD outperforms SOTAs on function-level and statement-level vulnerability detection across various target tasks, with an average increase of 40.9% in precision and 26.1% in recall. Notably, KF-GVD discovered 9 undisclosed vulnerabilities when employing on C/C++ open-source projects without ground truth. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Comparison of Different Optimization Techniques in AlexNet for Plant Disease Classification: SGD, Adadelta, ASGD and Adagrad 5

A Comparison of Different Optimization Techniques in AlexNet...

引用

5th International Conference on Smart Electronics and Communication, ICOSEC 2024

作者： Ponnila, P. Bazila Banu, A. Bannari Amman Institute of Technology Department of Information Technology Erode Tamilnadu Sathyamangalam India Kpr Institute of Engineering and Technology Department of Computer Science and Business Systems Tamilnadu Coimbatore India

ISBN: (纸本)9798331504403

Plant diseases are a biggest challenge for the farmers. Agriculture relies heavily on its production, and diagnosis of plant disease in an accurate manner will help in identification of the appropriate disease. If correct identification was not performed, there might be a significant financial loss that could have an impact on both farmers and the national economy. For resolving this issue, the AlexNet was employed, a convolutional neural network method. In this research, AlexNet is used to classify the diseases occurring in chilli, cauliflower and sunflower. This AlexNet model, with Stochastic Gradient Descent(SGD) algorithm optimization techniques had achieved 82% accuracy for the overall classification of the test images of chilli plant, 98% accuracy for the overall classification of the test images of cauliflower plant and 91% accuracy for the overall classification of the test images of sunflower plant;with Adadelta optimization 86%, 99% and 93% accuracy was achieved for the overall classification of the test pictures of chilli plant, cauliflower plant and sunflower plant respectively. After using the Asynchronous Stochastic Gradient(ASGD) optimization with AlexNet model, the accuracy obtained are 88% for the test images of chilli plant, 98% for the test images of cauliflower plant and 91% for the test images of sunflower plant;and finally, AdaGrad optimization with AlexNet achieved accuracy of 80% for the test images of chilli plant, 9 8% for the test images of cauliflower plant and 93% for the test images of sunflower plant. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

PLAME: Piecewise-Linear Approximate Measure for Additive Kernel SVM (Extended abstract) 40

PLAME: Piecewise-Linear Approximate Measure for Additive Ker...

引用

40th IEEE International Conference on Data engineering, ICDE 2024

作者： Chan, Tsz Nam Li, Zhe Leong Hou, U. Cheng, Reynold Shenzhen University Department of Computer Science and Software Engineering China Alibaba Cloud University of Macau State Key Laboratory of Internet of Things for Smart City Department of Computer and Information Science China The University of Hong Kong Department of Computer Science and Musketeers Foundation Institute of Data Science Hong Kong

ISBN: (纸本)9798350317152

Additive Kernel SVM has been extensively used in many applications, including human activity detection and pedestrian detection. Since training an additive kernel SVM model is very time-consuming, which is not scalable to largescale datasets, many efficient solutions have been developed in the past few years. However, most of the existing methods normally fail to achieve one of these three important conditions which are (1) low classification error, (2) low memory space, and (3) low training time. In order to simultaneously fulfill these three conditions, we develop the new piecewise-linear approximate measure (PLAME) for training additive kernel SVM models. Experimental results verify that this approach can achieve the best trade-off between accuracy, memory space, and training time compared with different types of state-of-the-art methods. © 2024 IEEE.

关键词： Additives

来源：评论

学校读者我要写书评

暂无评论

Secure Consumer-Centric Demand Response Management in Resilient Smart Grid as Industry 5.0 Application with Blockchain-Based Authentication

引用

IEEE Transactions on Consumer Electronics 2024年第1期70卷 1370-1379页

作者： Ayub, Muhammad Faizan Li, Xiong Mahmood, Khalid Shamshad, Salman Saleem, Muhammad Asad Omar, Marwan University of Electronic Science and Technology of China School of Computer Science and Engineering Sichuan Chengdu12599 China National Yunlin University of Science and Technology Graduate School of Intelligent Data Science Douliu64002 Taiwan The University of Lahore Department of Software Engineering Lahore54590 Pakistan Illinois Institute of Technology Information Technology and Management ChicagoIL60616 United States

The development of the industrial Internet of Things and smart grid networks has emphasized the importance of secure smart grid communication for the future of electric power transmission. However, the current deployment of smart meters and wireless communication methods presents several security concerns that must be addressed to ensure sustainable and resilient smart grid networks for Industry 5.0, with a focus on consumer-centric technologies. In this context, we propose a privacy-focused authentication solution for secure consumer-centric demand response management in smart grid networks, incorporating consumer electronics. Our proposed protocol utilizes blockchain-based authentication to enhance resistance against potential attacks and ensure the integrity of demand response data. Additionally, we have incorporated a PUF to prevent physical attacks. We have conducted informal and formal security analyses to verify the effectiveness of our proposed protocol against various threats and to confirm the session key security. Our comparison study has demonstrated that our protocol provides superior security and added functionalities, with lower computation and communication costs compared to similar protocols. Furthermore, we have measured the computational time for implementing the blockchain component of our protocol while considering the impact of varying numbers of block additions and transactions per block. © 1975-2011 IEEE.

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

Intrusion Detection Mechanism Using Deep Learning 5

Intrusion Detection Mechanism Using Deep Learning

引用

5th International Conference on IoT Based Control Networks and Intelligent systems, ICICNIS 2024

作者： Ananthi, P. Nirmaladevi, K. Naveen Kumar, S. Kongu Engineering College Department of Computer Technology-PG Tamilnadu Erode638060 India Madanapalle Institute of Technology & Science Department of Computer Applications India M.Sc Software Systems Kongu Engineering College Erode India

ISBN: (纸本)9798331518097

In today's connected and data-driven world, networks and digital systems need to be protected from malicious attacks. The effectiveness of conventional Intrusion Detection systems (IDS) in recognizing and impeding novel threats and intricate methodologies has its limits. This study introduces a new method for improving network security by using deep learning techniques in the design and development of an intrusion detection system (IDS). The suggested Intrusion Detection System uses deep learning techniques, specifically Long Short Term Memory (LSTM) and Convolutional Neural Networks (CNNs), to identify and react to known and new security threats. Data can be taught to be encoded into a lower-dimensional representation and then decoded back into its original form using autoencoders (AEs). Autoencoders are a cutting-edge intrusion detection system that can lower false positives and raise overall detection accuracy because of their capacity to automatically learn from and adjust to the ever-changing threat landscape. They enhance computer networks capacity to spot irregularities and possible security threats. With the deep learning model obtaining a high accuracy of 99% in real-world threat detection, the results show that deep learning is effective in developing intrusion detection systems. © 2024 IEEE.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

Facial Emotion Recognition using Deep Learning (FERDL) 25

Facial Emotion Recognition using Deep Learning (FERDL)

引用

25th International Multi Topic Conference, INMIC 2023

作者： Ramzan, Hafiz Arslan Sohaib, Ahmed Ramzan, Sadia Kalsum, Tehmina Aqeel, Muhammad Majeed, Mohsin Khwaja Fareed University of Engineering and Information Technology Institute of Computer and Software Engineering Rahim Yar Khan Pakistan National University of Modern Languages Department of Computer Science Multan Pakistan University of Engineering & Technology Department of Software Engineering Taxila Pakistan

ISBN: (纸本)9798350317701

Facial Emotion Recognition is one of the in-demand and rapidly growing research topics in the domain of computer Vision (CV) and artificial intelligence (AI). The ability to identify or detect human emotions from real-time facial expressions (FEs) has vast conceivable applications in different domains, such as sentiment analysis, human-computer interaction, human resource management, security, and human psychology. In this paper, a Convolutional Neural Network (CNN) based deep learning model is trained with haar-cascade classifier to recognize the real-time FEs. The suggested model is specially trained to categorize the FEs into one of the seven emotion categories, namely six basic emotions (sad, happy, angry, surprised, disgusted, fear) and a neutral emotion. It includes several convolutional layers, as well as fully connected neurons followed by max-pooling layers, and soft-max activation function with the corresponding seven classes. ReLU activation functions along with various kernels to enhance filtering depth, and extraction of facial features. FER-2013 dataset is used for experimentation purpose. To improve the classification performance and model accuracy, a data augmentation technique is used for rescaling and horizontal flipping. The proposed model outperforms the previous related works by achieving a validation accuracy of 71.96% and training accuracy above 90%, with fewer epochs. © 2023 IEEE.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Domain Adaptation for Action Recognition via Self-Ensembling and Conditional Embedding Alignment 24

Unsupervised Domain Adaptation for Action Recognition via Se...

引用

24th IEEE International Conference on Data Mining, ICDM 2024

作者： Ghosh, Indrajeet Chugh, Garvit Md Faridee, Abu Zaher Roy, Nirmalya University of Maryland Department of Information Systems Baltimore County United States Indian Institute of Technology Department of Computer Science and Engineering Jodhpur India Amazon United States

ISBN: (纸本)9798331506681

Recent advancements in deep learning-based wearable human action recognition (wHAR) have improved the capture and classification of complex motions, but adoption remains limited due to the lack of expert annotations and domain discrepancies from user variations. Limited annotations hinder the model's ability to generalize to out-of-distribution samples. While data augmentation can improve generalizability, unsupervised augmentation techniques must be applied carefully to avoid introducing noise. Unsupervised domain adaptation (UDA) addresses domain discrepancies by aligning conditional distributions with labeled target samples, but vanilla pseudo-labeling can lead to error propagation. To address these challenges, we propose μDAR, a novel joint optimization architecture comprised of three functions: (i) consistency regularizer between augmented samples to improve model classification generalizability, (ii) temporal ensemble for robust pseudo-label generation and (iii) conditional distribution alignment to improve domain generalizability. The temporal ensemble works by aggregating predictions from past epochs to smooth out noisy pseudo-label predictions, which are then used in the conditional distribution alignment module to minimize kernel-based class-wise conditional maximum mean discrepancy (kCMMD) between the source and target feature space to learn a domain invariant embedding. The consistency-regularized augmentations ensure that multiple augmentations of the same sample share the same labels;this results in (a) strong generalization with limited source domain samples and (b) consistent pseudo-label generation in target samples. The novel integration of these three modules in μDAR results in a range of 4-12% average macro-F1 score improvement over six state-of-the-art UDA methods in four benchmark wHAR datasets. © 2024 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Alzheimer's Disease Classification via Pre-trained Convolutional Neural Network Variants Utilizing Multi-level Thresholding based on Kapur's Entropy 2

Alzheimer's Disease Classification via Pre-trained Convoluti...

引用

2nd IEEE International Conference on Networks, Multimedia and information Technology, NMITCON 2024

作者： Mahendran, Nivedhitha Durai Raj Vincent, P.M. Srm Institute of Science and Technology School of Computing Technologies Department of Networking and Communications Kattankulathur Chennai India Vellore Institute of Technology School of Computer Science Engineering and Information Systems Vellore India

ISBN: (纸本)9798350372892

Neurological diseases are the disorders of brain and the nerves that run throughout the human body. There is a range of symptoms that arise as a result of chemical, structural, or electrical modifications in the regions of brain. One such neurological disease is the Alzheimer's Disease (AD), is considered as a slow and progressive disease that majorly affects the cognitive abilities of the individual. It is most common forms of dementia without any proper cure. However, early diagnosis and treatment might slow down the symptoms and improve the lifestyle of the individual. The most common techniques to identify the neurological disorders are imaging of the brain regions, namely, CT, MRI, PET, etc. The learning approaches, such as Deep Learning and Machine Learning are identified to be effective in analyzing these images to provide better accuracies in detecting the AD. In this study, four versions of Convolutional Neural Networks (CNN) are used to classify AD and Control cases. Further, the Kapur's thresholding technique is also implemented to segment the images for improving the classification accuracy of AD. From the results shown in the study, the inference shows that the VGG16 of CNN provides considerable improvement than the other deep learning models before and after the implementation of thresholding. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Dynamic Topology Management In Ad-Hoc Networks For Improved Performance 2

Dynamic Topology Management In Ad-Hoc Networks For Improved ...

引用

2nd IEEE International Conference on Advances in information Technology, ICAIT 2024

作者： Anakath, A.S. Kannadasan, R. Margarat, G. Simi Pandian, A. Pasumpon Antony Sibiya Varghese, V. Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Tamil Nadu Chennai India Department of Software Systems School of Computer Science and Engineering Vellore Institute of Technology VIT University Tamil Nadu Vellore India Department of Computer Science and Engineering New Prince Shri Bhavani College of Engineering and Technology Tamil Nadu Chennai India Department of Computer Science and Engineering CARE College of Engineering Trichy India Department of Artificial Intelligence and Data Science GRT Institute of Engineering and Technology Tamil Nadu Tiruttani India

ISBN: (纸本)9798350383867

Optimization in ad hoc networks is a highly specialized task because the structure of the network is loosely formed, and each node is independently responsible for its operation. In order to obtain a better result from this solely formed networks this report offer extensive Dynamic Topology Management (DTM) method. Dynamic routing protocols lager lines, dividend balancing, Life-saving administration, guarantee of reliability, cross-layer design, security aspects and the on-time monitoring and discovery of topology changes are the main objectives and steps to be accomplished. Real-time Location monitoring and forest structure discovery are made possible by employing adaptive routing protocols, namely Dynamic Source Routing(DSR) and Ad-Hoc On-Demand Distance Vector(AODV). Also, the algorithms are capable of making advance adjustments to routing routes and resource allocations drawn from the mobility patterns forecasting which is vital. Load balancing methods, which distribute traffic in a network evenly, enable the system to run efficiently and avoid problems caused by overloading. Algorithms that are energy-efficient and make the best use of energy resources having their topology's optimized are introduced to the streamlined energy distribution systems. These algorithms are focused on the node's energy level. Through fault tolerance techniques which recognize and handle failures in nodes and partitioned networks, connectedness activity is a continuous one. Through QoS perception, the network topology might automatically be tailored to conform to the given quality criteria, thus not restricting the choice for differing applications. The proposed could have defenses against adversary nodes and their potential DDoS attacks which will enhance network protection. Yet another advantage of an architectural style that promotes cross-layer collaboration is enhanced topology management which is the goal around which the functions of various levels of the protocol stack revolve

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Deep Ear Biometrics for Gender Classification 4th

Deep Ear Biometrics for Gender Classification

引用

4th International Conference on Communication, Devices and Computing (ICCDC 2023)

作者： Singh, Ritwiz Kashyap, Keshav Mukherjee, Rajesh Bera, Asish Chakraborty, Mamata Dalui Department of Computer Science and Engineering Haldia Institute of Technology WB Haldia India Department of Computer Science and Information Systems Birla Institute of Technology and Science Pilani Rajasthan Pilani Campus 333031 India Department of Computer Science and Engineering National Institute of Technology West Bengal Durgapur India

ISBN: (纸本)9789819927098

Human gender classification based on biometric features is a major concern for computer vision due to its vast variety of applications. The human ear is popular among researchers as a soft biometric trait, because it is less affected by age or changing circumstances and is non-intrusive. In this study, we have developed a deep convolutional neural network (CNN) model for automatic gender classification using the samples of ear images. The performance is evaluated using four cutting-edge pre-trained CNN models. In terms of trainable parameters, the proposed technique requires significantly less computational complexity. The proposed model has achieved 93% accuracy on the EarVN1.0 ear dataset. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Neural network models

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：