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 1-20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

Aspect Sentiment Triple Extraction Model Integrating with Syntactic Information 24

Aspect Sentiment Triple Extraction Model Integrating with Sy...

引用

5th International Conference on computer Science and Management Technology, ICCSMT 2024

作者： Zhao, Jingsheng Qu, Weilong Zheng, Jiashang Yu, Chen Zhu, Qiaoming School of Information and Control Engineering Qingdao University of Technology Shandong Qingdao China School of Computer Science and Technology Soochow University Jiangsu Suzhou China

ISBN: (纸本)9798400709999

The current mainstream aspect level sentiment analysis methods focus on modeling the relationship between word representations in text and their context, fail to fully utilize the syntactic structure information of the *** paper proposes a sentiment triplet extraction model IS-ASTE(Integrating with Syntactic information model for Aspect Sentiment Triplet Extraction, IS-ASTE).Firstly, the graph attention layer is employed to obtain contextual context ***, based on the syntactic dependency tree to obtain the text element's dependency, lexical and distance information, which is combined with the attention mechanism to augment the sentiment ***, graph convolutional neural networks are integrated to aggregate the feature information of nodes for sentiment *** experimental results indicate that, across four representative datasets, the F1 values are all better than those of the baseline model, demonstrating better performance in sentiment analysis. © 2024 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Invariant Set Analysis and control Synthesis of Multi-Equilibrium Switched Systems under control Constraints 13

Invariant Set Analysis and Control Synthesis of Multi-Equili...

引用

13th International Conference on Intelligent control and Information Processing, ICICIP 2025

作者： Han, Yuejiang Zhang, Lixian Ding, Yihang Dong, Yifei Liang, Ye School of Astronautics Harbin Institute of Technology Harbin China College of Computer and Control Engineering Northeast Forestry University Harbin China

ISBN: (纸本)9798331516147

This paper investigates the problems of invariant set analysis and control synthesis for multi-equilibrium switched systems under control constraints. A control strategy based on the invariant set method is proposed, accompanied by criteria formulated as linear matrix inequalities, ensuring that the state remains within the mode-dependent invariant sets. Furthermore, considering the control constraints in practical scenarios, a multi-equilibrium switched controller with control constraints (MESCC controller) is designed. Compared to existing works, the proposed method incorporates mode-dependent dwell time conditions and mode-dependent invariant sets, thereby offering enhanced design flexibility. An example of an aero-engine control system is presented to demonstrate the effectiveness and potential of the theoretical results. © 2025 IEEE.

关键词： control system synthesis

来源：评论

学校读者我要写书评

暂无评论

Research on indoor temperature and humidity coupling system based on fuzzy adaptive PD control 9

Research on indoor temperature and humidity coupling system ...

引用

2024 9th International Seminar on computer Technology, Mechanical and Electrical engineering, ISCME 2024

作者： Li, Hui Yang, Hao Zhang, Suli Miao, Jiahao School of Information and Control Engineering Jilin Institute of Chemical Technology Jilin Jilin132022 China School of Computer Technology and Engineering Changchun Institute of Technology Jilin Changchun130012 China

In this research, a fuzzy adaptive PD control approach is introduced for managing the coupled indoor temperature and humidity system. Initially, the mathematical framework of indoor temperature and humidity is analyzed. For the coupled system, a decoupling control strategy is introduced, which eliminates the mutual influence between temperature and humidity by incorporating decoupling components, thereby enabling the system to be equivalently controlled as two independent single-input single-output systems. Secondly, a fuzzy adaptive PD control method is employed to achieve parameter self-tuning. Finally, to assess the efficacy of the introduced control approach, the present study undertakes a simulation analysis, utilizing a medium-sized office as a case study. The outcomes of the simulation suggest that upon implementing the fuzzy adaptive PD control methodology, the modulation of indoor temperature and humidity achieves enhanced precision and swiftness, notably demonstrating greater robustness in the presence of disturbance signals. © 2025 Institute of Physics Publishing. All rights reserved.

关键词： Self tuning control systems

来源：评论

学校读者我要写书评

暂无评论

Non-contact all-optic OCT–PAM imaging with shared detection light

引用

Applied Optics 2025年第13期64卷 3431-3437页

作者： Ding, Ning Xiang, Ben Jiang, Huiwen Yue, Jinyun Yu, Yao Liu, Jian Zhao, Yuqian Luan, Jingmin Yang, Yanqiu Wang, Yi Ma, Zhenhe College of Information Science and Engineering Northeastern University Shenyang110819 China School of Control Engineering Northeastern University at Qinhuangdao Qinhuangdao066004 China Hebei Key Laboratory of Micro-Nano Precision Optical Sensing and Measurement Technology Qinhuangdao066004 China School of Computer and Communication Engineering Northeastern University at Qinhuangdao Qinhuangdao066004 China

We proposed a non-contact all-optic OCT–PAM dual-modal system based on a single detection light source. A homodyne low-coherence interferometer was adopted to detect the vibration induced by photoacoustic excitation. The optical coherence tomography (OCT) and photoacoustic microscopy (PAM) shared the same broadband light source for signal detection, thus simplifying the dual-modal system and making it compact. Using the proposed system, we performed absorption-scattering imaging of human hairs, measured the femoral artery and vein flow rates in mice, and imaged the auricle blood vessels in mice in vivo. The combination of OCT and PAM can provide complementary biological tissue information and offer a richer reference for the diagnosis of diseases. © 2025 Optica Publishing Group. All rights, including for text and data mining (TDM), Artificial Intelligence (AI) training, and similar technologies, are reserved.

关键词： Optical coherence tomography

来源：评论

学校读者我要写书评

暂无评论

Fully Distributed Model-Free Adaptive Sliding Mode control for MASs With Hybrid-Attacked Topology

引用

IEEE Transactions on Automation Science and engineering 2025年 22卷 12336-12346页

作者： Duan, Shitao Chen, Guangdeng Zhou, Qi Li, Hongyi Huang, Tingwen Guangdong University of Technology School of Automation Guangdong-Hong Kong Joint Laboratory for Intelligent Decision and Cooperative Control Guangdong Provincial Key Laboratory for Intelligent Decision and Cooperative Control Guangzhou510006 China Southwest University College of Electronic and Information Engineering Chongqing400715 China Shenzhen University of Advanced Technology Faculty of Computer Science and Control Engineering Shenzhen518055 China

A fully distributed model-free adaptive sliding mode control (MFASMC) strategy is proposed in this paper for unknown nonlinear multi-agent systems (MASs), in which topology networks are exposed to hybrid attacks consisting of denial-of-service (DoS) and false data injection attacks. Hybrid attacks in topology networks can result in neighbor information dropouts and inaccuracies among agents. First, the MASs with unknown dynamics are translated to equivalent linear data equations by the dynamic linearization technique. Second, the impact of neighbor information dropouts caused by DoS attacks is mitigated by a designed attack compensation mechanism, in which the compensation error is guaranteed to be bounded in the sense of mathematical expectation. Then, a fully distributed MFASMC algorithm, which does not depend on knowledge of the Laplacian matrix, is designed to improve the robustness of MASs with topology networks exposed to hybrid attacks, thus indirectly mitigating the impact of neighbor information inaccuracies. Finally, the consensus error is rigorously proved to be bounded in the sense of mathematical expectation, and the validity of the proposed strategy is confirmed by simulations. © 2004-2012 IEEE.

关键词： Network topology

来源：评论

学校读者我要写书评

暂无评论

Adaptive Particle Swarm Optimization of U-net Parameters for Marine Aquaculture Image Segmentation 13

Adaptive Particle Swarm Optimization of U-net Parameters for...

引用

13th International Conference on Intelligent control and Information Processing, ICICIP 2025

作者： Zhao, Jin Wang, Jun Han, Min Fan, Jianchao School of Computer Science and Engineering Chongqing Three Gorges University Chongqing China Department of Computer Science Department of Data Science City University of HongKong Hong Kong School of Control Science and Engineering Dalian University of Technology Dalian China

ISBN: (纸本)9798331516147

Marine aquaculture image segmentation plays a crucial role in managing aquatic resources and environmental protection. Traditional deep learning models rely on manual parameter tuning for image segmentation, which limits their efficiency and accuracy. This paper proposes an adaptive Particle Swarm Optimization (PSO) algorithm to optimize the parameters of the U-net model automatically. The algorithm dynamically adjusts the PSO parameters based on the population’s entropy and clustering metrics and employs a hill-climbing algorithm to address the issue of the PSO easily falling into local optima, enhancing the algorithm’s adaptability. The proposed algorithm updates and iterates particles to automatically find suitable model parameters. The remote sensing data used in this experiment were from the Yellow Sea aquaculture area near Dalian, China, with a segmentation accuracy of 91.8%. This method improves segmentation accuracy, reduces the burden of manual parameter tuning, and provides an effective solution for optimizing deep learning models. © 2025 IEEE.

关键词： Aquaculture

来源：评论

学校读者我要写书评

暂无评论

Interval prediction strategy for the remaining useful life of lithium-ion battery pack considering cell inconsistency

引用

Journal of Power Electronics 2025年 1-13页

作者： Pang, Xiaoqiong Li, Xiao Kim, Jonghoon Lee, Miyoung Zhao, Zhen Wen, Jie Zeng, Jianchao School of Computer Science and Technology North University of China Taiyuan China Department of Electrical Engineering Chungnam National University Daejeon Korea Republic of Institute of Network Technology Yantai China School of Electrical and Control Engineering North University of China Shanxi Taiyuan China

Accurate and robust remaining useful life (RUL) prediction of lithium-ion battery packs is critical for ensuring system operation reliability and safety. However, the inconsistency accelerates battery pack degradation, making RUL prediction challenging. Furthermore, traditional prediction methods offering specific values cannot meet varying safety and urgency requirements. Aiming at the above problems, this study proposed an RUL interval prediction scheme considering cell inconsistency for lithium-ion battery packs. First, the degradation and inconsistency information of the battery pack is captured by extracting health indicators from the measured data. Second, fuzzy information granulation is introduced to transform the time series data into fuzzy granules, providing upper and lower bounds for interval prediction modeling. Then, long short-term memory is employed for modeling, generating interval prediction results that support decision adjustment for different safety and emergency levels. Finally, the experiment results show that the proposed strategy can effectively cover the real value and achieve reliable interval prediction. It also indicates that the impact of inconsistency on the prediction results becomes more remarkable as the starting point moves back. With 60% training data, the interval prediction evaluation criterion P of the model with inconsistency information is approximately 43% higher than the model without. © The Author(s) under exclusive licence to The Korean Institute of Power Electronics 2025.

关键词： Battery Pack

来源：评论

学校读者我要写书评

暂无评论

Enhancing deformation characteristics prediction of coarse-grained soils with time-series generative adversarial network-based data augmentation and pre-training

引用

Frontiers of Structural and Civil engineering 2025年第3期19卷 396-410页

作者： Ying ZHANG Meng JIA Xuedong ZHANG Liping CAO Ziying AN Hongchao WANG Jinyu WANG School of Control and Computer Engineering North China Electric Power UniversityBeijing 102206China The State Key Laboratory of Simulation and Regulation of Water Cycle in River Basin China Institute of Water Resources and Hydropower ResearchBeijing 100038China

Coarse-grained soils are fundamental to major infrastructures like embankments,roads,and *** their deformation characteristics is essential for ensuring structural *** methods,such as triaxial compression tests and numerical simulations,face challenges like high costs,time consumption,and limited generalizability across different soils and *** address these limitations,this study employs deep learning to predict the volumetric strain of coarse-grained soils as axial strain changes,aiming to obtain the axial strain(ε_(a))-volumetric strain(ε_(v))curve,which helps derive key mechanical parameters like cohesion(c),and elastic modulus(E).However,the limited data from triaxial tests poses challenges for training deep learning *** propose using a Time-series Generative Adversarial Network(TimeGAN)for data ***,we apply feature importance analysis to assess the quality of the numerical augmented data,providing feedback for improving the TimeGAN *** further enhance model performance,we introduce the pre-training strategy to reduce bias between augmented and real *** results demonstrate that our approach effectively predictscurve,with the mean absolute error(MAE)of 0.2219 and the R^(2) of *** analysis aligns with established findings in soil mechanics,underscoring the potential of our method in engineering applications.

关键词： coarse-grained soils deformation characteristics TimeGAN data augmentation pre-training

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：