检索结果-内蒙古大学图书馆

A Novel Sparse Penalty for Singular Value Decomposition

Chinese Journal of Electronics 2017年第2期26卷 306-312页

作者： WANG Caihua LIU Juan MIN Wenwen QU Aiping The State Key Laboratory of Software Engineering School of ComputerWuhan University

Singular value decomposition(SVD) is a tool widely used in data denoising,matrix approximation,recommendation system,text mining and computer vision.A ma jority of applications prefer sparse singular vectors to capture inherent structures and patterns of the input data so that the results are *** present a novel penalty for SVD to achieve *** with the traditional penalties,the proposed penalty is scale,dimensional insensitive and bounded between 0 and 1,which are in favor of controlling *** by the penalty,we provide an efficient algorithm to pro ject a vector onto a given sparse level in O(n) expected *** efficient pro jection algorithm serve as a drudge for sparse SVD(SSVD).In experiments,SSVD is efficient and could capture the latent structures and patterns of the input data.

关键词： Sparse controllable projection(SCP) Sparse singular value decomposition(SSVD) Sparse low rank matrix approximation(SLRMA)

来源：评论

学校读者我要写书评

暂无评论

Continual Test-Time Adaptation for Single Image Defocus Deblurring via Causal Siamese Networks

arXiv

引用

arXiv 2025年

作者： Cui, Shuang Li, Yi Li, Jiangmeng Tang, Xiongxin Su, Bing Xu, Fanjiang Xiong, Hui National Key Laboratory of Space Integrated Information System Institute of Software Chinese Academy of Sciences University of Chinese Academy of Sciences China Beijing Key Laboratory of Big Data Management and Analysis Methods Gaoling School of Artificial Intelligence Renmin University of China China Department of Computer Science & Engineering the Hong Kong University of Science and Technology Hong Kong

Single image defocus deblurring (SIDD) aims to restore an all-in-focus image from a defocused one. Distribution shifts in defocused images generally lead to performance degradation of existing methods during out-of-distribution inferences. In this work, we gauge the intrinsic reason behind the performance degradation, which is identified as the heterogeneity of lens-specific point spread functions. Empirical evidence supports this finding, motivating us to employ a continual test-time adaptation (CTTA) paradigm for SIDD. However, traditional CTTA methods, which primarily rely on entropy minimization, cannot sufficiently explore task-dependent information for pixel-level regression tasks like SIDD. To address this issue, we propose a novel Siamese networks-based continual test-time adaptation framework, which adapts source models to continuously changing target domains only requiring unlabeled target data in an online manner. To further mitigate semantically erroneous textures introduced by source SIDD models under severe degradation, we revisit the learning paradigm through a structural causal model and propose Causal Siamese networks (CauSiam). Our method leverages large-scale pre-trained vision-language models to derive discriminative universal semantic priors and integrates these priors into Siamese networks, ensuring causal identifiability between blurry inputs and restored images. Extensive experiments demonstrate that CauSiam effectively improves the generalization performance of existing SIDD methods in continuously changing domains. Copyright © 2025, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Fed-GAN: Federated Generative Adversarial Network with Privacy-Preserving for Cross-Device Scenarios

引用

IEEE Transactions on Dependable and Secure Computing 2025年

作者： Han, Song Ding, Hongxin Zhao, Shuai Ren, Siqi Zeng, Shengke Xue, Mengqi Wang, Ruili Hangzhou City University School of Computer and Computing Science Hangzhou China Zhejiang Gongshang University School of Computer Science and Technology China Zhejiang Key Laboratory of Big Data and Future E-Commerce Technology Hangzhou China Xihua University School of Computer and Software Engineering Chengdu China Massey University School of Mathematical and Computational Science Auckland New Zealand

Huge amounts of data from various sources are substantial to dependable distributed machine learning, especially for trustworthy federated learning (FL). However, existing FL methods are difficult to collect enough data for training the global model more accurately, especially in cross-device scenarios. In this paper, we propose a new federated generative adversarial network empowered by differential privacy and knowledge transfer named Fed-GAN, which can be used to address the problem of data shortage and prevent generator leakage from resource-constrained devices, as well as generate high-quality synthetic data while ensuring strict DP guarantees. Different from other generative model methods, our Fed-GAN framework can achieve efficient and secure generative model training and limited permission for resource-constrained devices to prevent them from leaking or misusing the generator. In addition, we propose a pHash-KT method for our Fed-GAN framework, which selects potentially high-quality data through the knowledge of each client for improving the utility of synthetic data. Our FedGAN framework satisfies (Formula presented)-DP, and also has high resistance when number of adversaries is 10%-70% of the total number of clients. Extensive experiments demonstrate that our Fed-GAN framework not only generates high-quality synthetic data, but also provides strict DP guarantees, compared with other generative model methods. © 2004-2012 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Family of Mutually Uncorrelated Codes for DNA Storage Address Design

IEEE Transactions on Nanobioscience

引用

IEEE Transactions on Nanobioscience 2025年 PP卷 PP页

作者： Liu, Zhenlu Cao, Ben Shao, Qi Zheng, Yanfen Wang, Bin Zhou, Shihua Zheng, Pan Dalian University Key Laboratory of Advanced Design and Intelligent Computing Ministry of Education School of Software Engineering Dalian116622 China Dalian University of Technology School of Computer Science and Technology Dalian116024 China University of Canterbury Upper Riccarton Department of Accounting and Information Systems Christchurch8140 New Zealand

Deoxyribonucleic acid (DNA) has become an ideal medium for long-term storage and retrieval due to its extremely high storage density and long-term stability. But access efficiency is an existing bottleneck in DNA storage, especially the lack of high-quality random access address sequences. Therefore, in this paper, we report a series of approaches based on k-weakly mutually uncorrelated (k-WMU) codes to design the address sequence to improve the access efficiency of DNA storage. To address the problem of DNA sequences that are poorly scalable at the base level, we propose a 0-m-ruling coding scheme combined with k-WMU codes that can make address sequences avoid generating secondary structure with stem lengths ranging from 3 to 9. Based on the decoupled structure, We further extend the k-WMU codes with error correction function while satisfying combinatorial biological constraints. In order to investigate the performance of the designed address sequences for real-world applications, we perform simulation experiments based on thermodynamic properties and error correction capability as well as compared the minimum free energy (MFE), melting temperature (TM), and average decoding success rate (ADSR) with previous work. The results show that designed address sequences have a high MFE value and ADSR and a substantial reduction in TM-variance while satisfying the combinatorial biological constraints. As the quality of address sequences improves, this will help to achieve accurate random access as well as enhance the robustness of the DNA storage system. © 2002-2011 IEEE.

关键词： DNA sequences

来源：评论

学校读者我要写书评

暂无评论

TDiffSal: Text-Guided Diffusion Saliency Prediction Model for Images 27th

TDiffSal: Text-Guided Diffusion Saliency Prediction Model fo...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Zhang, Nana Xiong, Min Zhu, Dandan Zhu, Kun Zhai, Guangtao School of Computer Science and Technology Donghua University No. 2999 Renmin North Road Songjiang District Shanghai201620 China State Key Laboratory for Novel Software Technology Nanjing University Nanjing210093 China Institute of AI Education Shanghai East China Normal University No. 3663 Zhongshan North Road Putuo District Shanghai200333 China Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University No. 4800 Cao’an Highway Jiading District Shanghai201804 China Institute of Image Communication and Network Engineering Shanghai Jiao Tong University No. 800 Dongchuan Road Minhang District Shanghai200240 China

ISBN: (纸本)9783031781858

Existing visual saliency prediction methods mainly focus on single-modal visual saliency prediction, while ignoring the significant impact of text on visual saliency. To more comprehensively explore the influence of text on human attention in images, we propose a text-guided diffusion saliency prediction model, named TDiffSal. In specific, recent studies on stable diffusion models have shown promising performance in unifying tasks due to their inherent generalization ability. Inspired by this, a novel diffusion model for generalized visual-text saliency prediction is proposed, which formulates the prediction issue as a conditional generative task of the saliency map by employing input visual and text as the conditions. Meanwhile, we introduce a multi-head fusion module to effectively integrate text features and image features, which can efficiently guide the image denoising process and progressively refine the generated saliency map to make it semantically relevant to the text. Additionally, we employ an efficient pre-training strategy to enhance the robustness and generalization of the proposed model. We conduct extensive experiments on benchmark datasets to demonstrate its superior performance compared to other state-of-the-art methods. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Searching for an Accurate Robot Calibration via Improved Levenberg–Marquardt and Radial Basis Function System

引用

Journal of Field Robotics 2025年第0期

作者： Li, Zhibin Deng, Xun Chen, Tinghui Yang, Yuhang Chen, Linlin Yang, Xiwen Hu, Zhenzhen Hu, Lun Hu, Pengwei Li, Shuai Luo, Xin Xinjiang Technical Institute of Physics & Chemistry Chinese Academy of Sciences Urumqi China School of Software Engineering Chengdu University of Information Technology Chengdu China Dazhou Key Laboratory of Government Data Security Sichuan University of Arts and Science Dazhou China School of Software Engineering Xinjiang University Urumqi China Faculty of Information Technology and Electrical Engineering University of Oulu Oulu Finland VTT-Technical Research Centre of Finland Oulu Finland College of Computer and Information Science Southwest University Chongqing China

Robots are frequently utilized in manufacturing, aviation, and other industries, which enhance industrial production efficiency and quality. Specifically, robots perform high-precision tasks like welding, assembly and material handling, which reduce the intensity of manual labor in factories. In the logistics field, robots automatically sort and deliver goods, thereby speeding up supply chain operations. However, the prolonged operation of robots suffers from a decline in positioning accuracy, which makes them unable to satisfy task requirements. To address this challenging issue, this study designs an efficient calibration system integrating the Levenberg–Marquardt algorithm with fuzzy proportion integration differentiation controller and radial basis function neural network. The innovations of this method include: (1) integrating the fuzzy proportion integration differentiation controller into the updating rules of Levenberg–Marquardt algorithm, which further enhances the identification performance of kinematic errors;(2) adopting the radial basis function neural network to handle the robot dynamic errors, which addresses the complexity of dynamic error sources. Extensively experimental robot positioning points are gathered on an HSR JR680 robot, and then experimental validations are conducted by using the designed calibration system. The experiments indicate that the developed algorithm outperforms these existing advanced algorithms. © 2025 Wiley Periodicals LLC.

关键词： Radial basis function networks

来源：评论

学校读者我要写书评

暂无评论

BSDSGANet: Bidirectional Skip-stored Dual-Stream Gated Attention Network for multivariate time series classification

引用

Knowledge-Based Systems 2025年 318卷

作者： Yi, Yugen Zhao, Panpan Sheng, Hui Liu, Min Li, Shicheng Dai, Jiangyan Kong, Jun Qiao, Shaojie School of Software Research Center for Management Science and Engineering Jiangxi Provincial Key Laboratory of Intelligent Information Processing and Affective Computing Jiangxi Normal University Nanchang330022 China School of Artificial Intelligence and Big Data Jiangxi University of Technology Nanchang330098 China School of Computer Engineering Weifang University Weifang261061 China College of Information Science and Technology Northeast Normal University Changchun130117 China School of Software Engineering Chengdu University of Information Technology Chengdu610225 China

Echo State Networks (ESNs) are popular in Time Series Classification (TSC) due to their high-dimensional stochastic projections. However, due to the echo state at the current step is susceptible to the preceding step, the existing ESNs struggle to capture the long-term temporal dependencies. Furthermore, most of them only employ the unidirectional storage to capture the sequence information, which fails to fully consider the impact of the past and future moments simultaneously. To address these issues, we propose an advanced method called Bidirectional Skip-stored Dual-Stream Gated Attention Network (BSDSGANet) for Multivariate Time Series Classification (MTSC). Specifically, to roundly exploit the sequence information, a Bidirectional Skip-Stored Encoder (BSSE) is designed to enrich the temporal context by integrating past and future information. Then, a Dual-Stream Gated Attention Learner (DSGAL) mechanism is incroporated, which employs a Gated Recurrent Self-Attention (GRSA) module to learn the long-term dependencies and variable relationships. Meanwhile, a Multi-Scale Gated Attention (MSGA) module is developed to extract more important features from multiple scales. Extensive experiments conducted on eighteen benchmark MTSC datasets show that BSDSGANet outperforms five traditional methods and twelve deep learning models. © 2025 Elsevier B.V.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Attention-based UAV Trajectory Optimization for Wireless Power Transfer-assisted IoT Systems

arXiv

引用

arXiv 2025年

作者： Dong, Li Jiang, Feibo Peng, Yubo School of Computer Science Hunan University of Technology and Business Changsha410205 China Xiangjiang Laboratory Changsha410205 China Hunan Provincial Key Laboratory of Intelligent Computing and Language Information Processing Hunan Normal University Changsha410081 China School of Intelligent Software and Engineering Nanjing University Suzhou215163 China

Unmanned Aerial Vehicles (UAVs) in Wireless Power Transfer (WPT)-assisted Internet of Things (IoT) systems face the following challenges: limited resources and suboptimal trajectory planning. Reinforcement learning-based trajectory planning schemes face issues of low search efficiency and learning instability when optimizing large-scale systems. To address these issues, we present an Attention-based UAV Trajectory Optimization (AUTO) framework based on the graph transformer, which consists of an Attention Trajectory Optimization Model (ATOM) and a Trajectory lEarNing Method based on Actor-critic (TENMA). In ATOM, a graph encoder is used to calculate the self-attention characteristics of all IoTDs, and a trajectory decoder is developed to optimize the number and trajectories of UAVs. TENMA then trains the ATOM using an improved Actor-Critic method, in which the real reward of the system is applied as the baseline to reduce variances in the critic network. This method is suitable for high-quality and large-scale multi-UAV trajectory planning. Finally, we develop numerous experiments, including a hardware experiment in the field case, to verify the feasibility and efficiency of the AUTO framework. Copyright © 2025, The Authors. All rights reserved.

关键词： Antennas

来源：评论

学校读者我要写书评

暂无评论

UniUIR: Considering Underwater Image Restoration as An All-in-One Learner

arXiv

引用

arXiv 2025年

作者： Zhang, Xu Zhang, Huan Wang, Guoli Zhang, Qian Zhang, Lefei Du, Bo The Institute of Artificial Intelligence School of Computer Science Wuhan University Wuhan430072 China The Hubei Luojia Laboratory Wuhan China The National Engineering Research Center for Multimedia Software Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University Wuhan430072 China The School of Information Engineering Guangdong University of Technology Guangzhou510006 China The Horizon Robotics Beijing100083 China

Existing underwater image restoration (UIR) methods generally only handle color distortion or jointly address color and haze issues, but they often overlook the more complex degradations that can occur in underwater scenes. To address this limitation, we propose a Universal Underwater Image Restoration method, termed as UniUIR, considering the complex scenario of real-world underwater mixed distortions as an all-in-one manner. To decouple degradation-specific issues and explore the inter-correlations among various degradations in UIR task, we designed the Mamba Mixture-of-Experts module. This module enables each expert to identify distinct types of degradation and collaboratively extract task-specific priors while maintaining global feature representation based on linear complexity. Building upon this foundation, to enhance degradation representation and address the task conflicts that arise when handling multiple types of degradation, we introduce the spatial-frequency prior generator. This module extracts degradation prior information in both spatial and frequency domains, and adaptively selects the most appropriate task-specific prompts based on image content, thereby improving the accuracy of image restoration. Finally, to more effectively address complex, region-dependent distortions in UIR task, we incorporate depth information derived from a large-scale pre-trained depth prediction model, thereby enabling the network to perceive and leverage depth variations across different image regions to handle localized degradation. Extensive experiments demonstrate that UniUIR can produce more attractive results across qualitative and quantitative comparisons, and shows strong generalization than state-of-the-art methods. Copyright © 2025, The Authors. All rights reserved.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Anomaly-aware self-supervised feature learning for weakly supervised video anomaly detection

引用

computer Vision and Image Understanding 2025年 257卷

作者： Yang, Zhen Wang, Guodong Guo, Yuanfang Bao, Xiuguo Huang, Di State Key Laboratory of Complex and Critical Software Environment Beihang University Beijing100191 China School of Computer Science and Engineering Beihang University Beijing100191 China National Computer Network Emergency Response Technical Team/Coordination Center of China Beijing100029 China

Weakly supervised video anomaly detection (WSVAD) aims to achieve frame-level anomaly detection utilizing video-level labeled training data. Most of the WSVAD methods typically focus on developing various anomaly detectors and employ the models pretrained on general large-scale databases (e.g. Kinetics-400) for video feature extraction. However, such features are pretrained in a supervised manner to recognize significantly distinct human actions, making them sub-optimal for the WSVAD task, which involves detecting more complex events including subtle human and non-human motions. Unlike previous efforts, in this work, we address WSVAD from another fundamental perspective of feature learning, and considering the anomaly attributes, we propose an anomaly-aware self-supervised learning (SSL) based method. Specifically, we design a series of pretext tasks, including temporal order verification, speed prediction, arrow of time prediction, and abrupt change detection, to directly pretrain the model on the anomaly datasets, thus rendering features tailored for the WSVAD task. Moreover, to deal with the limitation of the existing methods which adapt the pretrained features to anomaly detection by utilizing top-k anomaly-scored samples, we present a hard instance mining strategy (HIMS) to additionally explore valuable cues from unused non-top-k ones, enhancing the discriminability between normal and abnormal instances. Experimental results clearly demonstrate that our method outperforms state-of-the-art counterparts on the UCF-Crime, ShanghaiTech and XD-Violence benchmark datasets, highlighting its effectiveness. © 2025 Elsevier Inc.

关键词： Self-supervised learning

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：