检索结果-南通市图书馆

Starlet:Network defense resource allocation with multi-armed bandits for cloud-edge crowd sensing in IoT

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Digital Communications and Networks 2024年第3期10卷 586-596页

作者： Hui Xia Ning Huang Xuecai Feng Rui Zhang Chao Liu College of computer science and technology Ocean University of ChinaQingdao266100China

The cloud platform has limited defense resources to fully protect the edge servers used to process crowd sensing data in Internet of *** guarantee the network's overall security,we present a network defense resource allocation with multi-armed bandits to maximize the network's overall ***,we propose the method for dynamic setting of node defense resource thresholds to obtain the defender(attacker)benefit function of edge servers(nodes)and ***,we design a defense resource sharing mechanism for neighboring nodes to obtain the defense capability of ***,we use the decomposability and Lipschitz conti-nuity of the defender's total expected utility to reduce the difference between the utility's discrete and continuous arms and analyze the difference ***,experimental results show that the method maximizes the defender's total expected utility and reduces the difference between the discrete and continuous arms of the utility.

关键词： Internet of things Defense resource sharing Multi-armed bandits Defense resource allocation

Meta label associated loss for fine-grained visual recognition

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

science China(Information sciences) 2024年第6期67卷 230-247页

作者： Yanchao LI Fu XIAO Hao LI Qun LI Shui YU School of computer science Nanjing University of Posts and Telecommunications School of Network Engineering Zhoukou Normal University School of computer science University of Technology Sydney

Recently, intensive attempts have been made to design robust models for fine-grained visual recognition, most notably are the impressive gains for training with noisy labels by incorporating a reweighting strategy into a meta-learning framework. However, it is limited to up or downweighting the contribution of an instance for label reweighting approaches in the learning process. To solve this issue, a novel noise-tolerant method with auxiliary web data is proposed. Specifically, first, the associations made from embeddings of well-labeled data with those of web data and back at the same class are measured. Next, its association probability is employed as a weighting fusion strategy into angular margin-based loss, which makes the trained model robust to noisy datasets. To reduce the influence of the gap between the well-labeled and noisy web data, a bridge schema is proposed via the corresponding loss that encourages the learned embeddings to be coherent. Lastly, the formulation is encapsulated into the meta-learning framework, which can reduce the overfitting of models and learn the network parameters to be noise-tolerant. Extensive experiments are performed on benchmark datasets, and the results clearly show the superiority of the proposed method over existing state-of-the-art approaches.

关键词： label associated loss weighting noisy samples fine-grained visual recognition noise-tolerant learning meta-learning

Understanding adversarial attacks on observations in deep reinforcement learning

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

science China(Information sciences) 2024年第5期67卷 69-83页

作者： You QIAOBEN Chengyang YING Xinning ZHOU Hang SU Jun ZHU Bo ZHANG department of computer science and technology Beijing National Research Center for Information Science and Technology Tsinghua-Bosch Joint Center for Machine Learning Institute for Artificial Intelligence Tsinghua University Peng Cheng Laboratory

Deep reinforcement learning models are vulnerable to adversarial attacks that can decrease the cumulative expected reward of a victim by manipulating its observations. Despite the efficiency of previous optimization-based methods for generating adversarial noise in supervised learning, such methods might not achieve the lowest cumulative reward since they do not generally explore the environmental ***, a framework is provided to better understand the existing methods by reformulating the problem of adversarial attacks on reinforcement learning in the function space. The reformulation approach adopted herein generates an optimal adversary in the function space of targeted attacks, repelling them via a generic two-stage framework. In the first stage, a deceptive policy is trained by hacking the environment and discovering a set of trajectories routing to the lowest reward or the worst-case performance. Next, the adversary misleads the victim to imitate the deceptive policy by perturbing the observations. Compared to existing approaches, it is theoretically shown that our adversary is strong under an appropriate noise level. Extensive experiments demonstrate the superiority of the proposed method in terms of efficiency and effectiveness, achieving state-of-the-art performance in both Atari and MuJoCo environments.

关键词： deep learning reinforcement learning adversarial robustness adversarial attack

Improving Video Watermarking through Galois Field GF(2^(4)) Multiplication Tables with Diverse Irreducible Polynomials and Adaptive Techniques

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

computers, Materials & Continua 2024年第1期78卷 1423-1442页

作者： Yasmin Alaa Hassan Abdul Monem S.Rahma department of computer science College of ScienceUniversity of BaghdadBaghdadIraq department of computer science College of ScienceAl-Maarif University CollegeAl-AnbarIraq

Video watermarking plays a crucial role in protecting intellectual property rights and ensuring content *** study delves into the integration of Galois Field(GF)multiplication tables,especially GF(2^(4)),and their interaction with distinct irreducible *** primary aim is to enhance watermarking techniques for achieving imperceptibility,robustness,and efficient execution *** research employs scene selection and adaptive thresholding techniques to streamline the watermarking *** selection is used strategically to embed watermarks in the most vital frames of the video,while adaptive thresholding methods ensure that the watermarking process adheres to imperceptibility criteria,maintaining the video's visual ***,careful consideration is given to execution time,crucial in real-world scenarios,to balance efficiency and *** Peak Signal-to-Noise Ratio(PSNR)serves as a pivotal metric to gauge the watermark's imperceptibility and video *** study explores various irreducible polynomials,navigating the trade-offs between computational efficiency and watermark *** parallel,the study pays careful attention to the execution time,a paramount consideration in real-world scenarios,to strike a balance between efficiency and *** comprehensive analysis provides valuable insights into the interplay of GF multiplication tables,diverse irreducible polynomials,scene selection,adaptive thresholding,imperceptibility,and execution *** evaluation of the proposed algorithm's robustness was conducted using PSNR and NC metrics,and it was subjected to assessment under the impact of five distinct attack *** findings contribute to the development of watermarking strategies that balance imperceptibility,robustness,and processing efficiency,enhancing the field's practicality and effectiveness.

关键词： Video watermarking galois field irreducible polynomial multiplication table scene selection adaptive thresholding

Japanese Sign Language Recognition by Combining Joint Skeleton-Based Handcrafted and Pixel-Based Deep Learning Features with Machine Learning Classification

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

computer Modeling in Engineering & sciences 2024年第6期139卷 2605-2625页

作者： Jungpil Shin Md.Al Mehedi Hasan Abu Saleh Musa Miah Kota Suzuki Koki Hirooka School of computer science and Engineering The University of AizuAizuwakamatsu965-8580Japan department of computer science&Engineering Rajshahi University of Engineering&TechnologyRajshahi6204Bangladesh

Sign language recognition is vital for enhancing communication accessibility among the Deaf and hard-of-hearing *** Japan,approximately 360,000 individualswith hearing and speech disabilities rely on Japanese Sign Language(JSL)for ***,existing JSL recognition systems have faced significant performance limitations due to inherent *** response to these challenges,we present a novel JSL recognition system that employs a strategic fusion approach,combining joint skeleton-based handcrafted features and pixel-based deep learning *** system incorporates two distinct streams:the first stream extracts crucial handcrafted features,emphasizing the capture of hand and body movements within JSL ***,a deep learning-based transfer learning stream captures hierarchical representations of JSL gestures in the second ***,we concatenated the critical information of the first stream and the hierarchy of the second stream features to produce the multiple levels of the fusion features,aiming to create a comprehensive representation of the JSL *** reducing the dimensionality of the feature,a feature selection approach and a kernel-based support vector machine(SVM)were used for the *** assess the effectiveness of our approach,we conducted extensive experiments on our Lab JSL dataset and a publicly available Arabic sign language(ArSL)*** results unequivocally demonstrate that our fusion approach significantly enhances JSL recognition accuracy and robustness compared to individual feature sets or traditional recognition methods.

关键词： Japanese Sign Language(JSL) hand gesture recognition geometric feature distance feature angle feature GoogleNet

Enhanced Differentiable Architecture Search Based on Asymptotic Regularization

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

computers, Materials & Continua 2024年第2期78卷 1547-1568页

作者： Cong Jin Jinjie Huang Yuanjian Chen Yuqing Gong School of computer science and technology Harbin University of Science and TechnologyHarbin150006China School of Automation Harbin University of Science and TechnologyHarbin150006China

In differentiable search architecture search methods,a more efficient search space design can significantly improve the performance of the searched architecture,thus requiring people to carefully define the search space with different complexity according to various *** rationalizing the search strategies to explore the well-defined search space will further improve the speed and efficiency of architecture *** this in mind,we propose a faster and more efficient differentiable architecture search method,***,we introduce a more efficient search space enriched by the introduction of two redefined convolution ***,we utilize a more efficient architectural parameter regularization method,mitigating the overfitting problem during the search process and reducing the error brought about by gradient ***,we introduce a natural exponential cosine annealing method to make the learning rate of the neural network training process more suitable for the search ***,group convolution and data augmentation are employed to reduce the computational ***,through extensive experiments on several public datasets,we demonstrate that our method can more swiftly search for better-performing neural network architectures in a more efficient search space,thus validating the effectiveness of our approach.

关键词： Differentiable architecture search allegro search space asymptotic regularization natural exponential cosine annealing

A Dual Discriminator Method for Generalized Zero-Shot Learning

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

computers, Materials & Continua 2024年第4期79卷 1599-1612页

作者： Tianshu Wei Jinjie Huang School of computer science and technology Harbin University of Science and TechnologyHarbin150006China School of Automation Harbin University of Science and TechnologyHarbin150006China

Zero-shot learning enables the recognition of new class samples by migrating models learned from semanticfeatures and existing sample features to things that have never been seen before. The problems of consistencyof different types of features and domain shift problems are two of the critical issues in zero-shot learning. Toaddress both of these issues, this paper proposes a new modeling structure. The traditional approach mappedsemantic features and visual features into the same feature space;based on this, a dual discriminator approachis used in the proposed model. This dual discriminator approach can further enhance the consistency betweensemantic and visual features. At the same time, this approach can also align unseen class semantic features andtraining set samples, providing a portion of information about the unseen classes. In addition, a new feature fusionmethod is proposed in the model. This method is equivalent to adding perturbation to the seen class features,which can reduce the degree to which the classification results in the model are biased towards the seen *** the same time, this feature fusion method can provide part of the information of the unseen classes, improvingits classification accuracy in generalized zero-shot learning and reducing domain bias. The proposed method isvalidated and compared with othermethods on four datasets, and fromthe experimental results, it can be seen thatthe method proposed in this paper achieves promising results.

关键词： Generalized zero-shot learning modality consistent discriminator domain shift problem feature fusion

Increasing the Maximum Capacity Path in a Network and Its Application for Improving the Connection Between Two Routers

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Tsinghua science and technology 2024年第3期29卷 753-765页

作者： Adrian M.Deaconu Javad Tayyebi department of Mathematics and computer science Transilvania University of BrasovBrasov 500036Romania department of Industrial Engineering Birjand University of TechnologyBirjand 000097Iran

Abstract:This paper addresses the problem of improving the optimal value of the Maximum Capacity Path(MCP)through expansion in a flexible network,and minimizing the involved *** only condition applied to the cost functions is to be non-decreasing *** is a non-restrictive condition,reflecting the reality in practice,and is considered for the first time in the ***,the total cost of expansion is a combination of max-type cost(e.g.,for supervision)and sum-type cost(*** building infrastructures,price of materials,price of labor,etc.).For this purpose,two types of strategies are combined:(l)increasing the capacity of the existing arcs,and(l)adding potential new *** different problems are introduced and *** the problems have immediate applications in Internet routing *** first one is to extend the network,so that the capacity of an McP in the modified network becomes equal to a prescribed value,therefore the cost of modifications is minimized.A strongly polynomial-time algorithm is deduced to solve this *** second problem is a network expansion under a budget constraint,so that the capacity of an McP is maximized.A weakly polynomial-time algorithm is presented to deal with *** the special case when all the costs are linear,a Meggido's parametric search technique is used to develop an algorithm for solving the problem in strongly polynomial *** new approach has a time complexity of O(n^(4)),which is better than the time complexity of O(n4 log(n)of the previously known method from literature.

关键词： Maximum Capacity Path(MCP) network expansion Internet routing polynomial-time algorithms

A New Industrial Intrusion Detection Method Based on CNN-BiLSTM

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

computers, Materials & Continua 2024年第6期79卷 4297-4318页

作者： Jun Wang Changfu Si Zhen Wang Qiang Fu College of computer science and technology Shenyang University of Chemical TechnologyShenyang110142China

Nowadays,with the rapid development of industrial Internet technology,on the one hand,advanced industrial control systems(ICS)have improved industrial production ***,there are more and more cyber-attacks targeting industrial control *** ensure the security of industrial networks,intrusion detection systems have been widely used in industrial control systems,and deep neural networks have always been an effective method for identifying cyber *** intrusion detection methods still suffer from low accuracy and a high false alarm ***,it is important to build a more efficient intrusion detection *** paper proposes a hybrid deep learning intrusion detection method based on convolutional neural networks and bidirectional long short-term memory neural networks(CNN-BiLSTM).To address the issue of imbalanced data within the dataset and improve the model’s detection capabilities,the Synthetic Minority Over-sampling Technique-Edited Nearest Neighbors(SMOTE-ENN)algorithm is applied in the preprocessing *** algorithm is employed to generate synthetic instances for the minority class,simultaneously mitigating the impact of noise in the majority *** approach aims to create a more equitable distribution of classes,thereby enhancing the model’s ability to effectively identify patterns in both minority and majority *** the experimental phase,the detection performance of the method is verified using two data *** results show that the accuracy rate on the CICIDS-2017 data set reaches 97.7%.On the natural gas pipeline dataset collected by Lan Turnipseed from Mississippi State University in the United States,the accuracy rate also reaches 85.5%.

关键词： Intrusion detection convolutional neural network bidirectional long short-term memory neural network multi-head self-attention mechanism