检索结果-内蒙古大学图书馆

Deep Learning for Cross-Domain Few-Shot Visual Recognition: A Survey

ACM Computing Surveys 2025年第8期57卷 1-37页

作者： Xu, Huali Zhi, Shuaifeng Sun, Shuzhou Patel, Vishal Liu, Li College of Computer Science Nankai University Tianjin China College of Electronic Science and Technology National University of Defense Technology Changsha China Faculty of Information Technology and Electrical Engineering University of Oulu Oulu Finland Johns Hopkins University Baltimore United States

While deep learning excels in computer vision tasks with abundant labeled data, its performance diminishes significantly in scenarios with limited labeled samples. To address this, Few-shot learning (FSL) enables models to perform the target tasks with very few labeled examples by leveraging prior knowledge from related tasks. However, traditional FSL assumes that both the related and target tasks come from the same domain, which is a restrictive assumption in many real-world scenarios where domain differences are common. To overcome this limitation, Cross-domain few-shot learning (CDFSL) has gained attention, as it allows source and target data to come from different domains and label spaces. This article presents the first comprehensive review of Cross-domain Few-shot Learning (CDFSL), a field that has received less attention compared to traditional FSL due to its unique challenges. We aim at providing both a position article and a tutorial for researchers, covering key problems, existing methods, and future research directions. The review begins with a formal definition of CDFSL, outlining its core challenges, followed by a systematic analysis of current approaches, organized under a clear taxonomy. Finally, we discuss promising future directions in terms of problem setups, applications, and theoretical advancements. © 2025 Copyright held by the owner/author(s).

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Nonsingular fast predefined time convergence sliding mode control for construction robot

引用

ISA Transactions 2025年 161卷 109-121页

作者： Yin, Chun-Wu Ding, Yun-Peng Sun, Hou-Jun College of Information and Control Engineering Xi'an University of Architecture and Technology Shaanxi Xi'an710055 China School of Computer Science Northwestern Polytechnical University Xi'an710072 China

To solve the practical engineering problem that the angle of a construction robot with an arbitrary initial value completely and accurately track the desired trajectory, this paper presents a control strategy for extending the desired trajectory, and a sliding mode controller with predefined time convergence is designed for an uncertain construction robot. Firstly, the special characteristics of the construction robot's trajectory tracking control are analyzed, an extended desired trajectory control strategy including the arrival phase and the precise tracking phase is proposed, and the design method of the extended desired trajectory in the arrival stage is given. Then, based on the improved non-singular fast sliding mode surface and a novel stability criterion of predefined time convergence, a predefined time convergence sliding mode controller based on a state observer was designed for the construction robot, which enables the angle tracking error of the construction robot to converge to zero within the specified time, ensures the angle completely and accurately tracking the original desired trajectory. Comparative simulation results show that the angle tracking error based on the proposed algorithm reaches 0.0004 rad, with an accuracy improvement of 33 %, and solves the problem of control torque chattering. © 2025 ISA

关键词： Sliding mode control

来源：评论

学校读者我要写书评

暂无评论

Data-driven output regulation control for constrained linear systems

引用

Science China(Information Sciences) 2025年第3期68卷 338-353页

作者： Chaoyu XIA Yi DONG Chaoli WANG Shengyuan XU Shanghai Research Institute for Intelligent Autonomous Systems Tongji University College of Electronic and Information Engineering Shanghai Research Institute for Intelligent Autonomous SystemsTongji University Department of Control Science and Engineering School of Optical-Electrical and Computer EngineeringUniversity of Shanghai for Science and Technology Department of Automation Nanjing University of Science and Technology

This study introduces a data-driven approach for state and output feedback control addressing the constrained output regulation problem in unknown linear discrete-time systems. Our method ensures effective tracking performance while satisfying the state and input constraints, even when system matrices are not available. We first establish a sufficient condition necessary for the existence of a solution pair to the regulator equation and propose a data-based approach to obtain the feedforward and feedback control gains for state feedback control using linear programming. Furthermore, we design a refined Luenberger observer to accurately estimate the system state, while keeping the estimation error within a predefined set. By combining output regulation theory, we develop an output feedback control strategy. The stability of the closed-loop system is rigorously proved to be asymptotically stable by further leveraging the concept of λ-contractive sets.

关键词： output regulation constrained system data-driven.

来源：评论

学校读者我要写书评

暂无评论

Frequency Attacks Based on Invertible Neural Networks

IEEE Transactions on Artificial Intelligence

引用

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

Interference Suppression and Jitter Elimination Ability-Based Adaption Tracking Guidance for Robotic Fishes

引用

IEEE/CAA Journal of Automatica Sinica 2025年第1期12卷 126-137页

作者： Dongfang Li Jie Huang Rob Law Xin Xu Limin Zhu Edmond Q.Wu IEEE the School of Electrical Engineering and Automation Fuzhou University the Key Laboratory of System Control and Information Processing Ministry of Education the School of Electrical Engineering and Automation Fuzhou University 5G+Industrial Internet Institute Fuzhou University the University of Macau the College of Mechatronics and Automation National University of Defense Technology the School of Mechanical Engineering Shanghai Jiao Tong University the Department of Computer Science and Engineering Shanghai Jiao Tong University

This work presents an adaptive tracking guidance method for robotic fishes. The scheme enables robots to suppress external interference and eliminate motion jitter. An adaptive integral surge line-of-sight guidance rule is designed to eliminate dynamics interference and sideslip issues. Limited-time yaw and surge speed observers are reported to fit disturbance variables in the model. The approximation values can compensate for the system's control input and improve the robots' tracking ***, this work develops a terminal sliding mode controller and third-order differential processor to determine the rotational torque and reduce the robots' run jitter. Then, Lyapunov's theory proves the uniform ultimate boundedness of the proposed method. Simulation and physical experiments confirm that the technology improves the tracking error convergence speed and stability of robotic fishes.

关键词： Adaptive integral surge line-of-sight approximation robotic fish speed observer

来源：评论

学校读者我要写书评

暂无评论

Enhancing Causal Discovery in Federated Settings with Limited Local Samples

Enhancing Causal Discovery in Federated Settings with Limi...

引用

International Workshop on Trustworthy Federated Learning, FL 2024

作者： Guo, Xianjie Yi, Liping Wu, Xiaohu Yu, Kui Wang, Gang School of Computer Science and Information Engineering Hefei University of Technology Hefei China College of Computer Science Nankai University Tianjin China National Engineering Research Center of Mobile Network Technologies Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9783031822391

Causal discovery from observational data is crucial for understanding complex systems, but traditional methods often require centralized data, conflicting with growing privacy concerns. Although federated causal discovery (FCD) has emerged as a solution, existing methods struggle when individual clients possess limited local samples. This paper introduces FedECD, a novel approach addressing causal discovery in federated settings with limited local samples. FedECD comprises two phases: 1) Federated Causal Skeleton Optimization and 2) Federated Causal Structure Refinement, both leveraging Bootstrapping techniques to enhance robustness and accuracy across distributed clients. Both phases employ a two-layer aggregation strategy: client-layer aggregates results from Bootstrapped sub-datasets within each client, while server-layer aggregates across all clients. The first phase uses weighted aggregation to iteratively remove false causal edges based on conditional independence tests. In contrast, the second phase utilizes majority voting to determine edge directions, ensuring robust estimation of the true causal structure. Extensive experiments on eight benchmark Bayesian network datasets demonstrate the superiority of FedECD over existing FCD methods, particularly with limited sample sizes. FedECD achieves an average improvement of 7.53% in the Ar_F1 score compared to the best baseline, addressing a critical challenge in FCD. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Federated learning

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

A stacking-based model for the spread of Botryosphaeria laricina

引用

Journal of Forestry Research 2025年第1期36卷 427-443页

作者： Hongwei Zhou Shibo Zhang Meng Xie Xiaodong Li Yifan Chen Wenhao Dai College of Computer and Control Engineering Northeast Forestry UniversityHarbin 150040People’s Republic of China Key Laboratory of National Forestry and Grassland Administration On Forest and Grassland Pest Monitoring and Warning Center for Biological Disaster Prevention and ControlNational Forestry and Grassland AdministrationShenyang 110034People’s Republic of China

Botryosphaeria laricina(larch shoot blight)was first identified in 1973 in Jilin Province,*** disease spread rapidly and caused considerable damage because its pathogenesis was unknown at the time and there were no effective controls or quarantine *** present,it shows a spreading trend,but most research can only conduct physiological analyses within a relatively short period,combining individual influencing ***,methods such as neural network models,ensemble learning algorithms,and Markov models are used in pest and disease prediction and ***,there may be fitting issues or inherent limitations associated with these *** study obtained *** data at the county level from 2003 to *** dataset was augmented using the SMOTE algorithm,and then algorithms such as XGBoost were used to select the significant features from a combined set of 12 features.A new stacking fusion model has been proposed to predict the status of *** model is based on random forest,gradient boosted decision tree,CatBoost and logistic regression *** accuracy,recall,specificity,precision,F_(1) value and AUC of the model reached 90.9%,91.6%,90.4%,88.8%,90.2%and 96.2%.The results provide evidence of the strong performance and stability of the *** is mainly found in the northeast and this study indicates that it is spreading *** means should be used promptly to prevent further damage and spread.

关键词： Botryosphaeria laricina Data augmentation Feature selection Stacking Status of spread

来源：评论

学校读者我要写书评

暂无评论

A Zero-Sequence Current Analysis Approach for Rotating Machinery Fault Diagnosis of Induction Motor Drivetrain Based on Sparse Learning

引用

IEEE Transactions on Power Electronics 2025年第7期40卷 9800-9810页

作者： Liu, Zhao-Hua Long, Jun-Jie Wei, Hua-Liang Liu, Kan Zhang, Ying-Jie Li, Xiao-Hua Hunan University of Science and Technology School of Information and Electrical Engineering Xiangtan411201 China The University of Sheffield School of Electrical and Electronic Engineering Department of Automatic Control and Systems Engineering SheffieldS1 3JD United Kingdom Hunan University College of Mechanical and Vehicle Engineering Changsha410082 China Hunan University College of Computer Science and Electronic Engineering Changsha410082 China

Fault diagnosis of rotating machinery driven by induction motors has received increasing attention. Current diagnostic methods, which can be performed on existing inverters or current transformers of three-phase induction machines, have become a more economical and reliable alternative to vibration diagnostic methods. Existing references mainly focus on the stator current analysis of single-phase, but most single-phase current analysis methods utilize only a fraction of the total information accessible in the three-phase system. Field experience shows that zero-sequence current (ZSC) has more distinctive fault characteristics compared to single-phase current, and therefore it is better to use in mechanical fault detection and diagnostic tasks. This article proposes a novel sparse learning-based method for ZSC analysis for induction rotating motor drive fault diagnosis. First, it elaborates and compares the effectiveness of ZSC in revealing fault characteristics compared with single-phase current by analyzing the fault diagnosis mechanism of ZSC. In addition, the method proposes a correlation entropy-enhanced sparse learning method for the problem of high-frequency noise and interference in the ZSC signals, so as to enhance the learning of the features of the noise-containing signals. The diagnostic efficacy of the proposed method has been verified through experiments on two real datasets. © 1986-2012 IEEE.

关键词： Induction motors

来源：评论

学校读者我要写书评

暂无评论

Multi-scale network for single image deblurring based on ensemble learning module

引用

Multimedia Tools and Applications 2025年第11期84卷 9045-9064页

作者： Wu, WenBo Pan, Yun Su, Na Wang, JingTao Wu, ShaoChuan Xu, ZeKun Yu, YouJian Liu, YaPeng State Key Laboratory of Media Convergence and Communication Communication University of China Beijing100024 China School of Computer and Cyberspace Security Communication University of China Beijing100024 China School of Electronics and Information Engineering Harbin Institute of Technology Harbin150080 China College of Computer and Control Engineering Northeast Forestry University Harbin150040 China School of Computer and Information Engineering Tianjin Chengjian University Tianjin300384 China School of Computer Science and Technology Anhui University Hefei230601 China

In this paper, we have identified two primary issues with current multi-scale image deblurring methods. On the one hand, the blurring scale is ignored. On the other hand, the context information of images is not fully utilized, and the spatial detail information of images is lost. To better solve the above problems, on top of that, this paper proposes a multi-scale network for single image deblurring. The designed network contains two parts: 1) Ensemble Learning Model: The Weak Learning Model in this module connects multiple weak learners in series to increase network depth. Moreover, through multi-scale feature ensemble module, the output features and intermediate features of different weak learner are integrated, and finally strong learner is used to enhance the features with semantic information at different scales;2) Spatial Feature Enhancement Module: This module can preserve spatial detail information of images and compensate for the disadvantages of losing image spatial details due to continuous downsampling in the Ensemble Learning Model structure. Our proposed deblurring model has been extensively evaluated on several benchmark datasets, and the model achieves superior performance compared to the most advanced deblurring methods. In particular, the peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) values are 33.57 dB and 0.965 on the GoPro dataset, respectively, with PSNR and SSIM improving by 0.49 dB and 0.002 over NAFNet, and 0.65 dB and 0.004 over Restormer. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Learning systems

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：