检索结果-内蒙古大学图书馆

Federated Class-Incremental Learning via Weighted Aggregation and Distillation

IEEE Internet of Things Journal 2025年第12期12卷 22489-22503页

作者： Wu, Feng Tan, Alysa Ziying Feng, Siwei Yu, Han Deng, Tao Zhao, Libang Chen, Yuanlu Soochow University School of Computer Science and Technology Suzhou215000 China Nanyang Technological University School of Computer Science and Engineering 639798 Singapore

Federated Class-Incremental Learning (FCIL) aims to design privacy-preserving collaborative training methods to continuously learn new classes from distributed datasets. In these scenarios, federated clients face the challenge of encountering new classes while being constrained by limited memory capacity, which can lead to catastrophic forgetting in the resulting global model. Existing FCIL approaches tend to overlook the challenges posed by the heterogeneity of dataset label distribution among clients, thereby constraining the generalization capacity of the global model they learn. Some of these methods also suffer from excessive computational burdens when addressing catastrophic forgetting problems. Furthermore, certain approaches are constrained to handling only straightforward data, posing significant difficulties in managing complex datasets and tackling more intricate scenarios. In this paper, we propose the Weighted Aggregation and Distillation-based FCIL (WAD-FCIL) method to address these limitations. To address data heterogeneity arising from class imbalance, we first introduce a task-aware client clustering method to identify clients with extreme class deviations before global model aggregation to eliminate potential impact on the global model. Then, we propose a multi-sampling weighted aggregation approach during the global FL model update that integrates knowledge from different clients and dynamically adjusts the weight of each client model to facilitate model update. To mitigate catastrophic forgetting, we propose a multi-model distillation strategy that involves selecting multiple teacher models for knowledge distillation. Extensive experiments comparing WAD-FCIL with ten state-of-the-art methods demonstrate that it significantly outperforms the baselines by 0.8%-3.2% in terms of average test accuracy on three representative benchmark datasets. © 2025 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Application of Face Detection and Recognition Algorithm Based on Deep Learning in Automatic Interception of Target Person Video Clips

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer science 2024年第10期51卷 1546-1559页

作者： Wang, Zhibiao Tao, Ye Liu, Xu Yu, Jiayi Cui, Wenhua School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan China

Current automatic segment extraction techniques for identifying target characters in videos have several limitations, including low accuracy, slow processing speeds, and poor adaptability to diverse scenes. This paper introduces an optimized algorithm to address these issues that enhance the RetinaFace and FaceNet models. We selected RetinaFace for face detection, employing MobileNetV1-0.25 as the backbone network and simplifying its Feature Pyramid Network (FPN) structure to boost detection speeds. Analysis of 460 images with a 720P resolution demonstrated an average speed improvement of 20.6%. For face recognition, we utilized FaceNet with MobileNetV3 as the backbone, augmenting its feature extraction capability by integrating four Receptive Field Block (RFB) structures and replacing the Squeeze-and-Excitation (SE) module with the Convolutional Block Attention Module (CBAM). Experimental results indicate that our enhancements elevate the maximum accuracy to 97%, outperforming the original model. Additionally, we integrated these refined algorithms and conducted disintegration experiments on segment extraction in 10 videos, evaluating various metrics. The findings show improvements in both precision and recall. We also compared our algorithm against the Dlib model;our system achieved an overall interception accuracy of 79.94%, surpassing Dlib’s 75.55%. This confirms the enhanced performance and feasibility of our proposed algorithm. © (2024), (International Association of Engineers). All rights reserved.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Image Enhancement via Associated Perturbation Removal and Texture Reconstruction Learning

引用

IEEE/CAA Journal of Automatica Sinica 2024年第11期11卷 2253-2269页

作者： Kui Jiang Ruoxi Wang Yi Xiao Junjun Jiang Xin Xu Tao Lu IEEE the School of Computer Science and Technology Harbin Institute of Technology Zhengzhou Research Institute Harbin Institute of Technology the School of Artificial Intelligence Jianghan University the School of Geodesy and Geomatics Wuhan University the Hubei Province Key Laboratory of Intelligent Information Processing and Real-Time Industrial System Wuhan University of Science and Technology the School of Computer Science and Engineering and also with the Hubei Province Key Laboratory of Intelligent Robot Wuhan Institute of Technology

Degradation under challenging conditions such as rain, haze, and low light not only diminishes content visibility, but also results in additional degradation side effects, including detail occlusion and color distortion. However, current technologies have barely explored the correlation between perturbation removal and background restoration, consequently struggling to generate high-naturalness content in challenging scenarios. In this paper, we rethink the image enhancement task from the perspective of joint optimization: Perturbation removal and texture reconstruction. To this end, we advise an efficient yet effective image enhancement model, termed the perturbation-guided texture reconstruction network(PerTeRNet). It contains two subnetworks designed for the perturbation elimination and texture reconstruction tasks, respectively. To facilitate texture recovery,we develop a novel perturbation-guided texture enhancement module(PerTEM) to connect these two tasks, where informative background features are extracted from the input with the guidance of predicted perturbation priors. To alleviate the learning burden and computational cost, we suggest performing perturbation removal in a sub-space and exploiting super-resolution to infer high-frequency background details. Our PerTeRNet has demonstrated significant superiority over typical methods in both quantitative and qualitative measures, as evidenced by extensive experimental results on popular image enhancement and joint detection tasks. The source code is available at https://***/kuijiang94/PerTeRNet.

关键词： Association learning attention mechanism image enhancement perturbation modeling

来源：评论

学校读者我要写书评

暂无评论

XRL-SHAP-Cache:an explainable reinforcement learning approach for intelligent edge service caching in content delivery networks

引用

science China(Information sciences) 2024年第7期67卷 46-71页

作者： Xiaolong XU Fan WU Muhammad BILAL Xiaoyu XIA Wanchun DOU Lina YAO Weiyi ZHONG School of Software Nanjing University of Information Science and Technology School of Computing and Communications Lancaster University School of Computing Technologies Royal Melbourne Institute of Technology State Key Laboratory for Novel Software Technology Nanjing University School of Computer Science and Engineering University of New South Wales Data 61 Commonwealth Scientific and Industrial Research Organization School of Computer Science Qufu Normal University

Content delivery networks(CDNs) play a pivotal role in the modern internet infrastructure by enabling efficient content delivery across diverse geographical regions. As an essential component of CDNs, the edge caching scheme directly influences the user experience by determining the caching and eviction of content on edge servers. With the emergence of 5G technology, traditional caching schemes have faced challenges in adapting to increasingly complex and dynamic network environments. Consequently, deep reinforcement learning(DRL) offers a promising solution for intelligent zero-touch network governance. However, the blackbox nature of DRL models poses challenges in understanding and making trusting decisions. In this paper,we propose an explainable reinforcement learning(XRL)-based intelligent edge service caching approach,namely XRL-SHAP-Cache, which combines DRL with an explainable artificial intelligence(XAI) technique for cache management in CDNs. Instead of focusing solely on achieving performance gains, this study introduces a novel paradigm for providing interpretable caching strategies, thereby establishing a foundation for future transparent and trustworthy edge caching solutions. Specifically, a multi-level cache scheduling framework for CDNs was formulated theoretically, with the D3QN-based caching scheme serving as the targeted interpretable model. Subsequently, by integrating Deep-SHAP into our framework, the contribution of each state input feature to the agent's Q-value output was calculated, thereby providing valuable insights into the decision-making process. The proposed XRL-SHAP-Cache approach was evaluated through extensive experiments to demonstrate the behavior of the scheduling agent in the face of different environmental *** results demonstrate its strong explainability under various real-life scenarios while maintaining superior performance compared to traditional caching schemes in terms of cache hit ratio, quality of service(QoS),a

关键词： deep reinforcement learning (DRL) explainable artificial intelligence (XAI) multi-level cache content delivery network (CDN) D3QN algorithm Deep-SHAP

来源：评论

学校读者我要写书评

暂无评论

Joint Sample Position Based Noise Filtering and Mean Shift Clustering for Imbalanced Classification Learning

引用

Tsinghua science and technology 2024年第1期29卷 216-231页

作者： Lilong Duan Wei Xue Jun Huang Xiao Zheng School of Computer Science and Technology Anhui University of TechnologyMaanshan 243032China Institute of Artificial Intelligence Hefei Comprehensive National Science CenterHefei 230088China

The problem of imbalanced data classification learning has received much *** classification algorithms are susceptible to data skew to favor majority samples and ignore minority *** weighted minority oversampling technique(MWMOTE)is an effective approach to solve this problem,however,it may suffer from the shortcomings of inadequate noise filtering and synthesizing the same samples as the original minority *** this end,we propose an improved MWMOTE method named joint sample position based noise filtering and mean shift clustering(SPMSC)to solve these ***,in order to effectively eliminate the effect of noisy samples,SPMsC uses a new noise filtering mechanism to determine whether a minority sample is noisy or not based on its position and distribution relative to the majority *** that MWMOTE may generate duplicate samples,we then employ the mean shift algorithm to cluster minority samples to reduce synthetic replicate ***,data cleaning is performed on the processed data to further eliminate class *** on extensive benchmark datasets demonstrate the effectiveness of SPMsC compared with other sampling methods.

关键词： imbalanced data classification oversampling noise filtering clustering

来源：评论

学校读者我要写书评

暂无评论

CBA: multi source fusion model for fast and intelligent target intention identification

引用

Journal of Systems Engineering and Electronics 2024年第2期35卷 406-416页

作者： WAN Shichang LI Qingshan WANG Xuhua LU Nanhua School of Computer Science and Technology Xidian UniversityXi’an 710071China School of Mathematics and Computer Application Shangluo UniversityShangluo 726000China

How to mine valuable information from massive multisource heterogeneous data and identify the intention of aerial targets is a major research focus at present. Aiming at the longterm dependence of air target intention recognition, this paper deeply explores the potential attribute features from the spatiotemporal sequence data of the target. First, we build an intelligent dynamic intention recognition framework, including a series of specific processes such as data source, data preprocessing,target space-time, convolutional neural networks-bidirectional gated recurrent unit-atteneion (CBA) model and intention recognition. Then, we analyze and reason the designed CBA model in detail. Finally, through comparison and analysis with other recognition model experiments, our proposed method can effectively improve the accuracy of air target intention recognition,and is of significance to the commanders’ operational command and situation prediction.

关键词： intention massive data deep network artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

Fault-tolerant Asynchronous Control for FM-II Model-based 2D Markov Jump Systems Under Actuator Failures and Mode Mismatches

引用

IAENG International Journal of Applied Mathematics 2024年第10期54卷 2083-2091页

作者： Han, Chengyi Zhang, Yu Chen, Qing Jiang, Taiping Zhou, Jianping School of Computer Science and Technology Anhui University of Technology Ma’anshan243032 China

This paper deals with fault-tolerant asynchronous control for Fornasini-Marchesini second model-based two-dimensional Markov jump systems under actuator failures and mode mismatches. The actuator failures are modeled as norm-bounded uncertainties and the mode mismatches between the plant and the designed controller are characterized by a hidden Markov model. Employing the Lyapunov direct method, a criterion is established to ensure the asymptotic mean square stability and H∞ noise suppression performance of the closed-loop system. Then, two design methods for the fault-tolerant asynchronous controller are proposed for the cases where the actuator fault matrix is known and unknown, respectively. The required controller gains can be determined through feasible solutions to the linear matrix inequalities. Finally, the effectiveness of these design methods is demonstrated by the Darboux equation. © (2024), (International Association of Engineers). All rights reserved.

关键词： Hidden Markov models

来源：评论

学校读者我要写书评

暂无评论

Multi-Scale Dynamic Hypergraph Convolutional Network for Traffic Flow Forecasting

引用

Journal of Shanghai Jiaotong university (science) 2024年 1-9页

作者： Dong, Zhaoxian Yu, Shuo Shen, Yanming School of Computer Science and Technology Dalian University of Technology Liaoning Dalian116024 China

This paper focuses on the problem of traffic flow forecasting, with the aim of forecasting future traffic conditions based on historical traffic data. This problem is typically tackled by utilizing spatio-temporal graph neural networks to model the intricate spatio-temporal correlations among traffic data. Although these methods have achieved performance improvements, they often suffer from the following limitations: These methods face challenges in modeling high-order correlations between nodes. These methods overlook the interactions between nodes at different scales. To tackle these issues, in this paper, we propose a novel model named multi-scale dynamic hypergraph convolutional network (MSDHGCN) for traffic flow forecasting. Our MSDHGCN can effectively model the dynamic higher-order relationships between nodes at multiple time scales, thereby enhancing the capability for traffic forecasting. Experiments on two real-world datasets demonstrate the effectiveness of the proposed method. © 2023, Shanghai Jiao Tong university.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Heuristic Search with Cut Point Based Strategy for Critical Node Problem

引用

Journal of computer science & technology 2024年第6期39卷 1328-1340页

作者： Zhi-Han Chen Shao-Wei Cai Jian Gao Shi-Ke Ge Chan-Juan Liu Jin-Kun Lin School of Computer Science and Technology University of Chinese Academy of SciencesBeijing 101408China Key Laboratory of System Software(Chinese Academy of Sciences)and State Key Laboratory of Computer Science Institute of SoftwareChinese Academy of SciencesBeijing 100190China College of Information Science and Technology Northeast Normal UniversityChangchun 130024China School of Computer Science and Technology Dalian University of TechnologyDalian 116024China SeedMath Technology Limited Beijing 100086China

The critical node problem(CNP)aims to deal with critical node identification in a graph,which has extensive applications in many *** CNP is a challenging task due to its computational complexity,and it attracts much attention from both academia and *** this paper,we propose a population-based heuristic search algorithm called CPHS(Cut Point Based Heuristic Search)to solve CNP,which integrates two main *** first one is a cut point based greedy strategy in the local search,and the second one involves the functions used to update the solution pool of the ***,a mutation strategy is applied to solutions with probability based on the overall average similarity to maintain the diversity of the solution *** are performed on a synthetic benchmark,a real-world benchmark,and a large-scale network benchmark to evaluate our *** with state-of-the-art algorithms,our algorithm has better performance in terms of both solution quality and run time on all the three benchmarks.

关键词： local search cut point heuristic search critical node problem

来源：评论

学校读者我要写书评

暂无评论

Joint multiple resource allocation for offloading cost minimization in IRS-assisted MEC networks with NOMA

引用

Digital Communications and Networks 2023年第3期9卷 613-627页

作者： Guang Chen Yueyun Chen Zhiyuan Mai Conghui Hao Meijie Yang Shuangshuang Han Liping Du School of Computer and Communication Engineering University of Science and Technology BeijingBeijing100083China Shunde Innovition School of Unversity of Science and Technology Beijing

Effective resource allocation can exploit the advantage of intelligent reflective surface(IRS)assisted mobile edge computing(MEC)***,it is challenging to balance the limited energy of MTs and the strict delay requirement of their *** this paper,in order to tackle the challenge,we jointly optimize the offloading delay and energy consumption of mobile terminals(MTs)to realize the delay-energy tradeoff in an IRS-assisted MEC network,in which non-orthogonal multiple access(NOMA)and multiantenna are applied to improve spectral *** achieve the optimal delay-energy tradeoff,an offloading cost minimization model is proposed,in which the edge computing resource allocation,signal detecting vector,uplink transmission power,and IRS phase shift coefficient are needed to be jointly *** optimization of the model is a multi-level fractional problem in complex fields with some coupled high dimension *** solve the intractable problem,we decouple the original problem into a computing subproblem and a wireless transmission subproblem based on the uncoupled relationship between different variable *** computing subproblem is proved convex and the closed-form solution is obtained for the edge computing resource ***,the wireless transmission subproblem is solved iteratively through decoupling the residual *** each iteration,the closed-form solution of residual variables is obtained through different successive convex approximation(SCA)*** verify the proposed algorithm can converge to an optimum with polynomial *** results indicate the proposed method achieves average saved costs of 65.64%,11.24%,and 9.49%over three benchmark methods respectively.

关键词： Mobile edge computing Intelligent reflective surface Task offloading Multiple resource allocation Non-orthogonal multiple access

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：