检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2025年第2期6卷 292-300页

作者： Shao, Ming-Wen Yang, Jian-Xin Meng, Ling-Zhuang Hu, Zhi-Yong China University of Petroleum School of Computer Science and Technology East China Qingdao266580 China China University of Petroleum College of Control Science and Engineering East China Qingdao266580 China

Adversarial attacks reveal the vulnerability of classifiers based on deep neural networks to well-designed perturbations. Most existing attack methods focus on adding perturbations directly to the pixel space. However, the perturbations generated by these methods may be easily perceived by humans. To alleviate the aforementioned problem, we propose a novel high-frequency attack based on invertible neural networks (HA-INN) that relies on INNs to adds perturbations to the high-frequency space of the image instead of the pixel space. In this way, we can fool the classifier while the perturbations are not easily detected by humans. Specifically, we introduce INNs to separate the high-frequency and low-frequency components of the image. And the low-frequency components are guaranteed to be reconstructed as the original image, while the high-frequency components are replaced with resampled high-frequency latent variables with additional adversarial information. Then, the low-frequency components and the high-frequency components with adversarial information are inversely fed into the INN to generate effective adversarial examples. Extensive experiments on two datasets (CIFAR-10 and CIFAR-100) show that our method can generate misleading and transferable cross-architectural adversarial examples with greatly reduced computational resource requirements. Under the white-box setting, the attack success rate of CIFAR10 and CIFAR100 is 99.8% and 99.74%, respectively. Further, under the black-box setting, the adversarial examples generated by our method are more effective than the other methods. © 2020 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Improving graph collaborative filtering with network motifs

引用

Neural Computing and Applications 2025年第16期37卷 9413-9432页

作者： Zhang, Yuqi Yu, Jian Liu, Zhizhong Wang, Guiling Nguyen, Minh Sheng, Quan Z. Wang, Nancy Department of Computer and Information Sciences Auckland University of Technology Auckland Central1010 New Zealand School of Computer and Control Engineering Yantai University Shandong Yantai264005 China School of Information Science and Technology North China University of Technology Beijing100144 China School of Computing Macquarie University SydneyNSW2109 Australia

Deep learning on graphs, specifically graph convolutional networks (GCNs), has exhibited exceptional efficacy in the domain of recommender systems. Most GCNs have a message-passing architecture that enables nodes to aggregate information from neighbours iteratively through multiple layers. This enables GCNs to learn from higher-order information, but the model does not allow for direct captions of the local structural patterns. Our rationale is to investigate the effectiveness of capturing such local patterns for graph-based collaborative filtering to enhance model’s learning ability per layer. This technique combines lower-order and higher-order interactions during layer-wise propagation. In this paper, we propose MotifGCN to aggregate both lower-order and higher-order information in each graph convolution layer. Specifically, we develop dedicated algorithms of generating motif adjacency matrices. The matrices are then used for motif-enhanced neighbourhood aggregation in each layer. As this paper focuses on recommender systems, MotifGCN is built on the basis of bipartite graphs. Our experiments on four real-world datasets show that MotifGCN has a superior performance compared to various state-of-the-art methods. © The Author(s) 2025.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

Phy-CoCo: Physical Constraint-Based Correlation Learning for Tropical Cyclone Intensity and Size Estimation 27

Phy-CoCo: Physical Constraint-Based Correlation Learning for...

引用

27th European Conference on Artificial Intelligence, ECAI 2024

作者： Yan, Hanting Mu, Pan Huang, Cheng Zhang, Jinglin Bai, Cong College of Computer Science Zhejiang University of Technology China School of Control Science and Engineering Shandong University China

ISBN: (纸本)9781643685489

Tropical Cyclone (TC) estimation aims to estimate various attributes of TC in real-time to alleviate and prevent disasters caused by violent TCs. As artificial intelligence technology advances, various deep learning-based multi-task estimation approaches have been proposed. However, most of them only focus on extracting common features of tasks, disregarding potential negative transfer and task interactions between different tasks. This paper is thus motivated to propose a Physical Constraint-based Correlation (Phy-CoCo) learning framework from the perspective of Multi-Task Learning (MTL). Specifically, for task-specific feature learning, we introduce Correlation Modeling (CoM) based on Centrally Expanded Pooling (CEP). Furthermore, for cross-task interaction, we propose a Multi-Domain Recurrent Convolution (MDRC) module to incorporate physical constraints into MTL. These physical constraints enable the transformation of different task features by simulating the physical relations among different attributes of TC. Lastly, in combination with a task-shared network that leverages the hybrid fusion of multi-modal data, our MTL framework accurately estimates various TC attributes. Extensive experiments conducted on our constructed dataset demonstrate that the proposed Phy-CoCo outperforms previous methods in TC estimation in terms of estimation error, verifying the potential of the physics-incorporated MTL model. © 2024 The Authors.

关键词： Tropical cyclone

来源：评论

学校读者我要写书评

暂无评论

Generating Adversarial Patterns in Facial Recognition with Visual Camouflage

引用

Journal of Shanghai Jiaotong University (Science) 2024年 1-12页

作者： Bao, Qirui Mei, Haiyang Wei, Huilin Lü, Zheng Wang, Yuxin Yang, Xin Key Laboratory of Social Computing and Cognitive Intelligence of Ministry of Education School of Control Science and EngineeringSchool of Computer Science and Technology Dalian University of Technology Liaoning Dalian116024 China Systems Engineering Institute Academy of Military Sciences Beijing100141 China

Deep neural networks, especially face recognition models, have been shown to be vulnerable to adversarial examples. However, existing attack methods for face recognition systems either cannot attack black-box models, are not universal, have cumbersome deployment processes, or lack camouflage and are easily detected by the human eye. In this paper, we propose an adversarial pattern generation method for face recognition and achieve universal black-box attacks by pasting the pattern on the frame of goggles. To achieve visual camouflage, we use a generative adversarial network (GAN). The scale of the generative network of GAN is increased to balance the performance conflict between concealment and adversarial behavior, the perceptual loss function based on VGG19 is used to constrain the color style and enhance GAN’s learning ability, and the fine-grained meta-learning adversarial attack strategy is used to carry out black-box attacks. Sufficient visualization results demonstrate that compared with existing methods, the proposed method can generate samples with camouflage and adversarial characteristics. Meanwhile, extensive quantitative experiments show that the generated samples have a high attack success rate against black-box models. © 2023, Shanghai Jiao Tong University.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Switching control for Identification Deception in Cyber-Physical Systems 63

Switching Control for Identification Deception in Cyber-Phys...

引用

63rd IEEE Conference on Decision and control, CDC 2024

作者： Mavridis, Christos Kanellopoulos, Aris Sandberg, Henrik Johansson, Karl Henrik Kth Royal Institute of Technology Division of Decision and Control School of Electrical Engineering and Computer Science Stockholm Sweden

ISBN: (纸本)9798350316339

We investigate the problem of deceiving a malicious agent employing an identification method to estimate the closed-loop dynamics of a cyber-physical system. In particular, we propose a moving target defense mechanism that utilizes stochastic switching between linear closed-loop dynamics to drive a linear system identification process of a potential adversary to sub-optimal solutions with non-vanishing error. We provide a statistical analysis of the induced identification error and show that it is not possible for any linear system identification method to reconstruct the average dynamics of a stochastic switched linear system. Finally, we utilize the theory of Markov jump linear systems to guarantee asymptotic stability of the switching system, and formulate the switching control problem as an optimization problem that guarantees stability while taking into account the trade-off between security and switching effort. Simulation results showcase the efficacy of the proposed approach in inducing identification error for the adversary using minimal switching. © 2024 IEEE.

关键词： Markov processes

来源：评论

学校读者我要写书评

暂无评论

Wireless Sensing for Material Identification: A Survey

引用

IEEE Communications Surveys and Tutorials 2024年第3期27卷 1598-1617页

作者： Chen, Yande Xu, Chongzhi Li, Kexin Zhang, Jia Guo, Xiuzhen Jin, Meng Zheng, Xiaolong He, Yuan Tsinghua University School of Software Beijing China Beijing University of Posts and Telecommunications School of Computer Science Beijing China Shanghai Jiao Tong University John Hopcroft Center for Computer Science Shanghai China Zhejiang University College of Control Science and Engineering Hangzhou China

As an application of fine-grained wireless sensing, RF-based material identification follows the paradigm of RF computing that fetches the information during RF signal propagation. Specifically, the RF signal accesses the objects' material-related information and carries the information with its electromagnetic properties. With a variety of important applications, research on RF-based material identification has gained significant progress in recent years. However, several fundamental problems remain insufficiently studied, such as the sensing models, signal processing approaches, performance and future extensions. This paper presents the first comprehensive survey of RF-based material identification. According to the basic sensing model used for sensing, we propose a taxonomy to classify the existing works into two categories: reflection-based and penetration-based. The works in each category are further grouped by the type of RF signals used, with elaborated discussion of the detailed approaches and the common challenges. We provide a framework that benchmarks the performance of the existing works, followed by a thorough discussion of future extensions. © 1998-2012 IEEE.

关键词： II-VI semiconductors

来源：评论

学校读者我要写书评

暂无评论

Data Security Storage Mechanism Based on Blockchain Network

引用

computers, Materials & Continua 2023年第3期74卷 4933-4950页

作者： Jin Wang WeiOu Wenhai Wang RSimon Sherratt Yongjun Ren Xiaofeng Yu School of Computer&Communication Engineering Changsha University of Science&TechnologyChangsha410004China School of Control Science and Engineering Zhejiang UniversityHangzhou310058China School of Systems Engineering The University of ReadingReadingRG66AYUK School of Computer Engineering Research Center of Digital Forensics of Ministry of EducationNanjing University of Information Science&TechnologyNanjing210044China School of Business Nanjing UniversityNanjing210093China

With the rapid development of information technology,the development of blockchain technology has also been deeply *** performing block verification in the blockchain network,if all transactions are verified on the chain,this will cause the accumulation of data on the chain,resulting in data storage *** the same time,the security of data is also challenged,which will put enormous pressure on the block,resulting in extremely low communication efficiency of the *** traditional blockchain system uses theMerkle Tree method to store *** verifying the integrity and correctness of the data,the amount of proof is large,and it is impossible to verify the data in batches.A large amount of data proof will greatly impact the verification efficiency,which will cause end-to-end communication delays and seriously affect the blockchain system’s stability,efficiency,and *** order to solve this problem,this paper proposes to replace the Merkle tree with polynomial commitments,which take advantage of the properties of polynomials to reduce the proof size and communication *** realizing the ingenious use of aggregated proof and smart contracts,the verification efficiency of blocks is improved,and the pressure of node communication is reduced.

关键词： Blockchain cryptographic commitment smart contract data storage

来源：评论

学校读者我要写书评

暂无评论

Combining Multi-Scale Feature Fusion and Attention Mechanism to Achieve Pulmonary Nodule Classification 17

Combining Multi-Scale Feature Fusion and Attention Mechanism...

引用

17th International Congress on Image and Signal Processing, BioMedical engineering and Informatics, CISP-BMEI 2024

作者： He, Naixin Dong, Zhengyu Li, Lin Wang, Jinxia Ma, Chaoqing School of Computer and Control Engineering Yantai University China Yantaishan Hospital Affiliated Binzhou Medical University China

ISBN: (纸本)9798331507398

Lung cancer is the deadliest form of cancer, with most cases originating from small malignant nodules. However, the early symptoms of malignant nodules are not obvious, which can lead to misdiagnosis and delay the optimal treatment time. Therefore, differentiating between benign and malignant lung nodules is crucial for the early diagnosis and treatment of lung cancer. The advent of convolutional neural networks has made it possible to achieve rapid classification of pulmonary nodules. Most classification networks use a convolution kernel size of 3, but due to the limited receptive field, the classification of benign and malignant nodules remains inaccurate. To address this, we use larger convolution kernels to increase the receptive field, capture more detailed feature information, and enhance the ability of model to represent complex patterns and details. Furthermore, we fuse features from multiple scales to mitigate the impact of scale differences. We also employ an attention mechanism to obtain global information, highlight important areas, and suppress irrelevant parts. We use the Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) dataset to evaluate our model. Compared to other classification methods, our approach achieved superior performance, with an accuracy of 0.9855 and an F1score of 0.9901. © 2024 IEEE.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Smart Diagnosis of Active Systems 28th

Smart Diagnosis of Active Systems

引用

28th International Conference on Knowledge Based and Intelligent information and engineering Systems, KES 2024

作者： Lamperti, Gianfranco Zhao, Xiangfu Department of Information Engineering University of Brescia Brescia25123 Italy School of Computer and Control Engineering Yantai University Yantai264005 China

Automated diagnosis has always been a challenging task to AI. When model-based diagnosis is adopted, a model of the system is required in order to generate a set of diagnoses based on a collection of observations, where a diagnosis is a set of faulty components or, more generally, a set of faults ascribed to components. An active system (AS) is an asynchronous, distributed discrete-event system, whose model consists of a topology (how components are connected to one another), and a communicating automaton for each component (the mode in which a component reacts to events). A problem afflicting all model-based approaches to diagnosis is a possibly large number of diagnoses explaining the observations, which may jeopardize the task of a diagnostician in charge of monitoring the system, owing to the cognitive overload raised by an overwhelming number of faulty scenarios to examine. This is exacerbated in critical application domains, where, under uncertain conditions, an artificial agent is supposed to perform recovery actions in real-time, even in the order of milliseconds, to possibly restore the system. To make diagnosis of ASs viable in critical, real-time application domains, a Smart Diagnosis Engine is presented, which is grounded on two heuristics: (1) if a diagnosis δ is a superset of a diagnosis δ′, then δ is ignored (minimality);(2) if the cardinality (number of faults) of a diagnosis δ is lower than the cardinality of a diagnosis δ′, then δ is generated before δ′ (sorting). Consequently, the diagnosis output consists in a sequence of minimal diagnoses that are generated in ascending order by cardinality. As indicated by the experimental results, the overall improvement is twofold: most likely diagnoses are generated upfront, thereby supporting real-time recovery actions;also, the abductive search in the behavior space of the AS is reduced considerably, owing to the pruning of the trajectories that will not generate minimal diagnoses, thereby resulting in an

关键词： Finite automata

来源：评论

学校读者我要写书评

暂无评论

The Estimation Method of Multi-target Parameters for OTFS Radar Sensing

The Estimation Method of Multi-target Parameters for OTFS Ra...

引用

2024 Photonics and Electromagnetics Research Symposium, PIERS 2024

作者： Chang, Quanran Wang, Jixuan Tian, Tuanwei Deng, Hao Du, Xiaolin Henan University School of Physics and Electronics Kaifeng475001 China Yantai University School of Computer and Control Engineering Yantai264005 China

ISBN: (纸本)9798350375909

In high mobility beyond 5G communications, orthogonal time frequency space (OTFS) is a promising alternative to orthogonal frequency division multiplexing. In addition to the superior communication performance, OTFS is also a natural choice for radar sensing. In this work, we investigate the estimation problem with multi-target parameters for the OTFS-based radar sensing system. To begin with, we recap the fundamental concepts in OTFS and present the input-output relation of OTFS symbols in the delay-Doppler domain. Then, we conceive a multitarget parameter estimation algorithm for range and velocity parameter estimation. Finally, numerical results demonstrate that the proposed algorithm can get the multi-target parameters with a high estimation accuracy. © 2024 IEEE.

关键词： Radar target recognition

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：