检索结果-内蒙古大学图书馆

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Yakun Chang Yeliduosi Xiaokaiti Yujia Liu Bin Fan Zhaojun Huang Tiejun Huang Boxin Shi Institute of Information Science School of Computer Science Beijing Jiaotong University Beijing Key Laboratory of Advanced Information Science and Network Technology National Key Laboratory for Multimedia Information Processing School of Computer Science Peking University National Engineering Research Center of Visual Technology School of Computer Science Peking University National Key Lab of General AI School of Intelligence Science and Technology Peking University

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

The spiking cameras offer the benefits of high dynamic range (HDR), high temporal resolution, and low data redundancy. However, reconstructing HDR videos in high-speed conditions using single-bit spikings presents challenges due to the limited bit depth. Increasing the bit depth of the spikings is advantageous for boosting HDR performance, but the readout efficiency will be decreased, which is unfavorable for achieving a high frame rate (HFR) video. To address these challenges, we propose a readout mechanism to obtain rolling-mixed-bit (RMB) spikings, which involves inter-leaving multi-bit spikings within the single-bit spikings in a rolling manner, thereby combining the characteristics of high bit depth and efficient readout. Furthermore, we introduce RMB-Net for reconstructing HDR and HFR videos. RMB-Net comprises a cross-bit attention block for fusing mixed-bit spikings and a cross-time attention block for achieving temporal fusion. Extensive experiments conducted on synthetic and real-synthetic data demonstrate the superiority of our method. For instance, pure 3 -bit spikings result in 3 times of data volume, whereas our method achieves comparable performance with less than 2% increase in data volume.

关键词： computer vision Computational modeling Redundancy Cameras Boosting Data models Pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Social-enhanced recommendation using graph-based contrastive learning 25

Social-enhanced recommendation using graph-based contrastive...

引用

25th IEEE International Conferences on High Performance Computing and Communications, 9th International Conference on Data Science and Systems, 21st IEEE International Conference on Smart City and 9th IEEE International Conference on Dependability in Sensor, Cloud and Big Data Systems and Applications, HPCC/DSS/SmartCity/DependSys 2023

作者： Xue, Peng Gao, Qian Fan, Jun Qilu University of Technology Shandong Academy of Sciences Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Shandong Jinan250014 China Qilu University of Technology Shandong Academy of Sciences Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Shandong Jinan250353 China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Shandong Jinan250014 China China Telecom Digital Intelligence Techonology Co Ltd Shandong Jinan250101 China

ISBN: (纸本)9798350330014

The social network-based recommendation model use social network information to mitigate data sparsity issues and improve the accuracy of recommendation models. However, In most social network-based recommendation algorithms, the social neighbors' contributions are difficult to distinguish from the central user's, neglecting hidden correlations in social infor-mation. In order to solve this problem, this paper presents a graph-based contrastive learning framework for social-enhanced recommendation (SoGCLR) that utilizes implicit social information captured by a new social relation attention mechanism, en-rich user representations, and improve model robustness through graph-based contrastive learning. Specifically, the paper captures the degree of contribution of each neighbor in the social graph to the central user through a social relation attention layer, thus obtaining hidden correlations in social information, and further integrates this with user information in the user-item interaction graph to enrich user representations. In addition, the paper incorporates graph-based contrastive learning into the recommendation task using cross-layer contrastive learning, calculating contrastive loss and mapping nodes with similar but different exposure rates to nearby regions to mitigate exposure bias issues. Results from Ciao and Epinions demonstrate that SoGCLR reduces RMSE and MAE by 1.33% to 1.84% compared with baseline models. © 2023 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

A Legal Judgment Prediction Model Based on BERT, Attention, and Graph Convolutional network

A Legal Judgment Prediction Model Based on BERT, Attention, ...

引用

International Conference on Tools for Artificial Intelligence (ICTAI)

作者： Binxia Yang Guibin Chen Xudong Luo School of Computer Science and Engineering Guangxi Normal University Guilin Guangxi China Key Lab of Education Blockchain and Intelligent Technology of Ministry of Education Guilin Guangxi China Guangxi Key Lab of Multi-Source Information Mining & Security Guilin Guangxi China

ISBN: (数字)9798331527235

ISBN: (纸本)9798331527242

Legal judgment prediction is one of essential tasks in legal practice. However, most existing models of legal judgment prediction struggle with long documents due to limitations in position encoding and pre-trained language models' capabilities. To address these issues, we propose a hybrid model for legal judgment prediction. The model uses BERT to captures deep semantic information of a legal text, position encoding to preserve its text structure, a graph convolutional network to analyse complex relationships between the entities in the text, and a decoupled attention mechanism to fuse the textual content and positional information. Our experiments show that the proposed model excels in legal judgment prediction tasks, outperforming existing technologies in accuracy and recall, with significant advantages in handling lengthy legal texts.

关键词： Analytical models Attention mechanisms Law Graph convolutional networks Text categorization Semantics Bidirectional control Predictive models Encoding Context modeling

来源：评论

学校读者我要写书评

暂无评论

Outlier Detection in Eigenvalue Spaces Based on Spectral Analysis of KNN Graphs

SSRN

引用

SSRN 2023年

作者： Li, Jia Wang, Chenxu Wang, Xiaochun Wang, Xiali School of Software Engineering Xi’an Jiaotong University Xi’an710049 China MOE Key Lab of Intelligent Network and Network Security Xi’an Jiaotong University China Department of Computer Science Chang’an University Xi’an China

Outlier detection is a critical method in data mining as it has a wide range of practical use in various fields. Spectral clustering has attracted much attention for its advantage of coping with the curse of dimensionality in high-dimensional spaces. This paper presents a new spectral analysis-based outlier detection (SAOD) framework. We first construct the $k$ nearest neighbors ($k$NN) graphs of each sample and compute the eigenvalues of the Laplacian matrices of these graphs. Our findings indicate that the distribution of eigenvalues for outliers differs from that of inliers. Based on this observation, we employ the multivariate Gaussian kernel density estimation to empirically estimate the eigenvalues distribution of a given dataset. We then introduce a new outlier factor, based on the eigenvalue probabilities of samples to assess their outlierness degree. The experimental results, obtained from both synthetic and real-world datasets, show that our spectral analysis-based outlier detection method outperforms the state-of-the-art approaches. © 2023, The Authors. All rights reserved.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

ASDIA: An Adversarial Sample to Preserve Privacy Program in Federated Learning

引用

IEEE Transactions on Dependable and Secure Computing 2025年

作者： Tian, Shujuan Tan, Yi Wang, Han Liu, Haolin Li, Zhetao Xiangtan University School of Computer Science School of Cyberspace Security Xiangtan China Xiangtan University Hunan International Scientific and Technological Cooperation Base of Intelligent network Xiangtan411105 China Xiangtan University Key Laboratory of Hunan Province for Internet of Things and Information Security Xiangtan411105 China Department of Electrical Engineering and Computer Science University of Kansas 1520 W 15th St LawrenceKS United States Jinan University College of Information Science and Technology Guangzhou510632 China

Federated learning enables training across multiple entities while ensuring data security and the effectiveness of knowledge dissemination. Despite its benefits, it remains susceptible to privacy breaches by both external and internal adversaries, who may exploit data or model parameters to glean sensitive participant information or disrupt the training process, thus compromising participant privacy and security. This paper proposes a novel methodology, Adversarial Samples for Defense Inference Attack (ASDIA), aimed at dual protection of data privacy and model robustness within federated learning through adversarial samples and gradient reconstruction. ASDIA includes gradient processing approach before uploading: initially identifying privacy-sensitive gradient, followed by the injection of well-calibrated noise to these gradients. This method not only obfuscates the adversary's classification demarcations but also aids in model performance recovery, all the while maintaining computational efficiency. ASDIA reduces the efficacy of attacks to near-random guessing levels and shows better balance between the model utility and privacy protection compared to the most advanced defense strategies. Additionally, regarding model performance, ASDIA proves its merit across diverse datasets under overfitting and non-overfitting scenarios. © 2004-2012 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

A Real-time Cutting Model Based on Finite Element and Order Reduction

引用

computer Systems Science & Engineering 2022年第10期43卷 1-15页

作者： Xiaorui Zhang Wenzheng Zhang Wei Sun Hailun Wu Aiguo Song Sunil Kumar Jha Wuxi Research Institute Nanjing University of Information Science&TechnologyWuxi214100China Engineering Research Center of Digital Forensics Ministry of EducationJiangsu Engineering Center of Network MonitoringSchool of Computer and SoftwareNanjing University of Information Science&TechnologyNanjing210044China School of Automation Nanjing University of Information Science&TechnologyNanjing210044China State Key Laboratory of Bioelectronics Jiangsu Key Lab of Remote Measurement and ControlSchool of Instrument Science and EngineeringSoutheast UniversityNanjing210096China IT Fundamentals and Education Technologies Applications University of Information Technology and Management in RzeszowRzeszow Voivodeship100031Poland

Telemedicine plays an important role in Corona Virus Disease 2019(COVID-19).The virtual surgery simulation system,as a key component in telemedicine,requires to compute in ***,this paper proposes a realtime cutting model based on finite element and order reduction method,which improves the computational speed and ensure the real-time *** proposed model uses the finite element model to construct a deformation model of the virtual ***,a model order reduction method combining proper orthogonal decomposition and Galerkin projection is employed to reduce the amount of deformation *** addition,the cutting path is formed according to the collision intersection position of the surgical instrument and the lesion area of the virtual ***,the Bezier curve is adopted to draw the incision outline after the virtual lung has been ***,the simulation system is set up on the PHANTOM OMNI force haptic feedback device to realize the cutting simulation of the virtual *** results show that the proposed model can enhance the real-time performance of telemedicine,reduce the complexity of the cutting simulation and make the incision smoother and more natural.

关键词： Virtual surgery cutting model finite element model model order reduction Bezier curve

来源：评论

学校读者我要写书评

暂无评论

A High-Dimensional Temporal Data Publishing Method Based on Dynamic Bayesian networks and Differential Privacy

A High-Dimensional Temporal Data Publishing Method Based on ...

引用

International Joint Conference on Neural networks (IJCNN)

作者： Yaxin Wang Zhen Zhang Heng Qian Yongchao Gao Qiuyue Wang Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Massive high-dimensional data generated by the Internet typically contains sensitive privacy information. Protecting data privacy while maintaining utility has become a pressing challenge. We propose a novel high-dimensional temporal data publishing method leveraging dynamic Bayesian networks and differential privacy. Initially, a dynamic Bayesian network is constructed, utilizing mutual information filtering of data. Subsequently, we calculate the Coherent Neighborhood Propinquity for each node within the network to determine edge sensitivity and establish a privacy budget. Noise is then strategically added to attribute data in accordance with the sensitivity and privacy budget requirements, ensuring the dataset complies with ε-differential privacy standards. Experimental results demonstrate that the data availability performance of the SMAP dataset (Soil Moisture Active Passive) surpasses that of competing algorithms while providing an equivalent level of privacy protection. Hence, our method significantly enhances data availability without compromising differential privacy protection.

关键词： Differential privacy Privacy Sensitivity Publishing Time series analysis Dynamic scheduling Bayes methods

来源：评论

学校读者我要写书评

暂无评论

Grouped Federated Meta-Learning for Privacy-Preserving Rare Disease Diagnosis

Grouped Federated Meta-Learning for Privacy-Preserving Rare ...

引用

International Joint Conference on Neural networks (IJCNN)

作者： Xinru Song Feng Wu Zongchao Xie Longxiang Gao Lei Cui Youyang Qu Shujun Gu Key Laboratory of Computing Power Network and Information Security Ministry of EducationShandong Computer Science Center Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China School of Computer Science University of Technology Sydney Australia Faculty of Science University of Melbourne Australia

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Federated learning (FL) has been widely applied in medical field, which allows clients to collaboratively train global models without sharing local data. Nevertheless, the diversity and scarcity of samples from rare diseases may result in a decline in the performance of local models on client-side due to using a singular global model. Moreover, direct transmission of local models or parameters will likely lead to user privacy violations. To solve these problems, we propose a Grouped Federated Meta-Learning (GrFML) method to improve the performance of local personalization models while protecting data privacy. Specifically, we first utilize a self-attention mechanism to extract partial features from the client’s local data, which are uploaded to the server (medical data is susceptible to perturbation and data integrity, thus this process does not expose the private data). The server groups clients with similar features based on these extracted features. Then, multiple meta-models are trained on these groups and distributed back to the clients to enhance the performance of the client’s local models. Furthermore, during the FL process, we introduce dynamic perturbation to the uploaded gradients based on the model’s test accuracy to protect their privacy. Typically, the perturbation magnitude is directly proportional to the model’s test accuracy. Extensive experiments shown that the GrFML model significantly improves client personalization model accuracy and achieves a good privacy-utility trade-off.

关键词： Metalearning Privacy Data privacy Accuracy Perturbation methods Feature extraction Data models

来源：评论

学校读者我要写书评

暂无评论

Fault Diagnosis for Rolling Bearing Based on RF-LSTM

Fault Diagnosis for Rolling Bearing Based on RF-LSTM

引用

2021 CAA Symposium on Fault Detection, Supervision, and Safety for Technical Processes, SAFEPROCESS 2021

作者： Li, Qiuting Wang, Xiuqing Yang, Yunpeng Wang, Ruiyi Lv, Feng Hebei Normal University College of Computer and Cyber Security Shijiazhuang050024 China Hebei Provincial Key Laboratory of Network & Information Security Shijiazhuang China Hebei Provincial Engineering Research Center for Supply Chain Big Data Analytics & Data Security Shijiazhuang China

ISBN: (纸本)9781665401159

In the era of big data, it is of great significance to study data-driven fault diagnostic methods for equipment in complex industrial environments. In this paper, a data-driven fault diagnostic method for rolling bearing based on RF-LSTM is proposed, which combines random forest (RF) algorithm with long and short term memory (LSTM) networks. RF is used to select sensor data's features, which have great influence on rolling bearing fault diagnosis, and LSTM is used for prediction rolling bearing's working states. The proposed method is used for fault diagnosis of the Case Western Reserve University (CWRU) bearing dataset and the experimental results validated the effectiveness of the proposed RF-LSTM model. Experimental results also show that the performance of the RFLSTM model is better than that of RF, LSTM, RF-RF, PCA-RF, PCA-LSTM models. © 2021 IEEE.

关键词： Failure analysis

来源：评论

学校读者我要写书评

暂无评论

RobustPFL: Robust Personalized Federated Learning

引用

IEEE Transactions on Dependable and Secure Computing 2025年

作者： Chen, Guorong Wang, Wei Wu, Yufang Li, Chao Xu, Guangquan Ji, Shouling Li, Tao Shen, Meng Han, Yufei Xi'an Jiaotong University Ministry of Education Key Lab for Intelligent Networks and Network Security Xi'an China Beijing Jiaotong University 3 Shangyuancun Beijing Key Laboratory of Security and Privacy in Intelligent Transportation Beijing100044 China Tianjin University Tianjin Key Laboratory of Advanced Networking College of Intelligence and Computing Tianjin300350 China Zhejiang University College of Computer Science and Technology Hangzhou310027 China ITAI Haihe Lab Tianjin China Nankai University College of Computer Science Tianjin300350 China Beijing Institute of Technology School of Cyberspace Security Beijing100081 China Peng Cheng Laboratory Cyberspace Security Research Center Shenzhen518066 China INRIA Bretagne Rennes35042 France

Conventional federated learning (FL) coordinated by a central server focuses on training a global model and protecting the privacy of clients' training data by storing it locally. However, the statistical heterogeneity hinders the global model from adapting to the non-IID distributions among clients. Moreover, untrusted and unreliable central servers and malicious clients may compromise model integrity and availability, thus degrading the robustness of FL. To address these challenges, we present RobustPFL, a decentralized personalized federated learning (PFL) approach that combines α-based Layer-position Normalized Similarity (α-LNS) and local collaborative training to improve personalized performance while utilizing a blockchain-based committee mechanism to coordinate the aggregation process, thereby achieving high personalized accuracy and robustness. Extensive experiments show that our RobustPFL approach outperforms multiple algorithms, including Local training, FedAvg, FedReptile, Per-FedAvg, FedBN, and SPFL, on MNIST, CIFAR10, EMNIST, and N-BaIoT datasets in four non-IID settings. We also evaluate RobustPFL's effectiveness against attacks - poisoning attacks and free-riding attacks. Particularly, for three prevalent poisoning attacks (backdoor, label flipping, and model poisoning attacks), we compare non-defensive (FedAvg) and defensive (Krum, trimmed mean, Bulyan, FedBN, FLAME, and FangTrmean) methods with our proposed RobustPFL. The results show that our approach achieves significant defensive effects. © 2004-2012 IEEE.

关键词： Federated 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：