检索结果-内蒙古大学图书馆

Geometric prior guided hybrid deep neural network for facial beauty analysis

CAAI Transactions on Intelligence technology 2024年第2期9卷 467-480页

作者： Tianhao Peng Mu Li Fangmei Chen Yong Xu David Zhang The School of Computer Science and Technology Guizhou UniversityGuiyangChina Department of Automation Moutai InstituteRenhuaiGuizhouChina The School of Computer Science and Technology Harbin Institute of TechnologyShenzhenShenzhenChina The Information and Communication Engineering Department Dalian Minzu UniversityDalianChina The School of Data Science The Chinese University of Hong KongShenzhenShenzhenChina

Facial beauty analysis is an important topic in human *** may be used as a guidance for face beautification applications such as cosmetic *** neural networks(DNNs)have recently been adopted for facial beauty analysis and have achieved remarkable ***,most existing DNN-based models regard facial beauty analysis as a normal classification *** ignore important prior knowledge in traditional machine learning models which illustrate the significant contribution of the geometric features in facial beauty *** be specific,landmarks of the whole face and facial organs are introduced to extract geometric features to make the *** by this,we introduce a novel dual-branch network for facial beauty analysis:one branch takes the Swin Transformer as the backbone to model the full face and global patterns,and another branch focuses on the masked facial organs with the residual network to model the local patterns of certain facial ***,the designed multi-scale feature fusion module can further facilitate our network to learn complementary semantic information between the two *** model optimisation,we propose a hybrid loss function,where especially geometric regulation is introduced by regressing the facial landmarks and it can force the extracted features to convey facial geometric *** performed on the SCUT-FBP5500 dataset and the SCUT-FBP dataset demonstrate that our model outperforms the state-of-the-art convolutional neural networks models,which proves the effectiveness of the proposed geometric regularisation and dual-branch structure with the hybrid *** the best of our knowledge,this is the first study to introduce a Vision Transformer into the facial beauty analysis task.

关键词： deep neural networks face analysis face biometrics image analysis

来源：评论

学校读者我要写书评

暂无评论

Central Attention Mechanism for Convolutional Neural Networks

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer science 2024年第10期51卷 1642-1648页

作者： Geng, Y.X. Wang, L. Wang, Z.Y. Wang, Y.G. School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China Automation Design Institute Metallurgical Engineering Technology Co. Ltd. Dalian116000 China

Model performance has been significantly enhanced by channel attention. The average pooling procedure creates skewness, lowering the performance of the network architecture. In the channel attention approach, average pooling is used to collect feature information to provide representative values. By leveraging the central limit theorem, we hypothesize that the strip-shaped average pooling operation will generate a one-dimensional tensor by considering the spatial position information of the feature map. The resulting tensor, obtained through average pooling, serves as the representative value for the features, mitigating skewness during the process. By incorporating the concept of the central limit theorem into the channel attention operation process, this study introduces a novel attention mechanism known as the"Central Attention Mechanism (CAM)." Instead of directly using average pooling to generate channel representative values, the central attention approach employs star-stripe average pooling to normalize multiple feature representative values into a single representative value. In this way, strip-shaped average pooling can be utilized to collect data and generate a one-dimensional tensor, while star-stripe average pooling can provide feature representative values based on different spatial directions. To generate channel attention for the complementary input features, the activation of the feature representation value is performed for each channel. Our attention approach is flexible and can be seamlessly incorporated into various traditional network structures. Through rigorous testing, we demonstrate the effectiveness of our attention strategy, which can be applied to a wide range of computer vision applications and outperforms previous attention techniques. © (2024), (International Association of Engineers). All rights reserved.

关键词： Tensors

来源：评论

学校读者我要写书评

暂无评论

AttrLeaks on the Edge: Exploiting Information Leakage from Privacy-Preserving Co-inference

引用

Chinese Journal of Electronics 2023年第1期32卷 1-12页

作者： WANG Zhibo LIU Kaixin HU Jiahui REN Ju GUO Hengchang YUAN Wei School of Cyber Science and Engineering Wuhan University School of Cyber Science and Technology Zhejiang University Department of Computer Science and Technology Tsinghua University

Collaborative inference(co-inference) accelerates deep neural network inference via extracting representations at the device and making predictions at the edge server, which however might disclose the sensitive information about private attributes of users(e.g.,race). Although many privacy-preserving mechanisms on co-inference have been proposed to eliminate privacy concerns, privacy leakage of sensitive attributes might still happen during inference. In this paper, we explore privacy leakage against the privacy-preserving co-inference by decoding the uploaded representations into a vulnerable form. We propose a novel attack framework named AttrL eaks, which consists of the shadow model of feature extractor(FE), the susceptibility reconstruction decoder,and the private attribute classifier. Based on our observation that values in inner layers of FE(internal representation) are more sensitive to attack, the shadow model is proposed to simulate the FE of the victim in the blackbox scenario and generates the internal ***, the susceptibility reconstruction decoder is designed to transform the uploaded representations of the victim into the vulnerable form, which enables the malicious classifier to easily predict the private attributes. Extensive experimental results demonstrate that AttrLeaks outperforms the state of the art in terms of attack success rate.

关键词： Deep learning Privacy Collaboration Transforms Feature extraction Prediction algorithms Iron

来源：评论

学校读者我要写书评

暂无评论

Enhancing Offensive Language Detection with Data Augmentation and Knowledge Distillation

引用

Research 2024年第2期6卷 29-40页

作者： Jiawen Deng Zhuang Chen Hao Sun Zhexin Zhang Jincenzi Wu Satoshi Nakagawa Fuji Ren Minlie Huang The CoAl group DCSTInstitute for Artificial InteligenceState Key Lab of Intelligent Technology and SystemsBeijing National Research Center for Information Science and TechnologyTsinghua UniversityBeijing 100084China Graduate School of Information Science&Technology The University of Tokyo.Tokyo 1138654Japan School of Computer Science and Engineering University of Electronic Science and Technology of ChinaChengduSichuan611731China

Offensive language detection has received important attention and plays a crucial role in promoting healthy communication on social platforms,as well as promoting the safe deployment of large language *** data is the basis for developing detectors;however,the available offense-related dataset in Chinese is severely limited in terms of data scale and coverage when compared to English *** significantly affects the accuracy of Chinese offensive language detectors in practical applications,especially when dealing with hard cases or out-of-domain *** alleviate the limitations posed by available datasets,we introduce AugCOLD(Augmented Chinese Offensive Language Dataset),a large-scale unsupervised dataset containing 1 million samples gathered by data crawling and model ***,we employ a multiteacher distillation framework to enhance detection performance with unsupervised *** is,we build multiple teachers with publicly accessible datasets and use them to assign soft labels to *** soft labels serve as a bridge for knowledge to be distilled from both AugCOLD and multiteacher to the student network,i.e.,the final offensive *** conduct experiments on multiple public test sets and our well-designed hard tests,demonstrating that our proposal can effectively improve the generalization and robustness of the offensive language detector.

关键词： COLD dealing limitations

来源：评论

学校读者我要写书评

暂无评论

A new intelligent cross-domain routing method in SDN based on a proposed multiagent reinforcement learning algorithm

引用

International Journal of Intelligent Computing and Cybernetics 2024年第2期17卷 330-362页

作者： Miao Ye Lin Qiang Huang Xiao Li Wang Yong Wang Qiu Xiang Jiang Hong Bing Qiu School of Information and Communication Guilin University of Electronic TechnologyGuilinChina School of Computer and Information Security Guilin University of Electronic TechnologyGuilinChina School of Computer Science and Technology Xidian UniversityXianChina School of Optoelectronic Egineering Guilin University of Electronic TechnologyGuilinChina

Purpose–A cross-domain intelligent software-defined network(SDN)routing method based on a proposed multiagent deep reinforcement learning(MDRL)method is ***/methodology/approach–First,the network is divided into multiple subdomains managed by multiple local controllers,and the state information of each subdomain is flexibly obtained by the designed SDN multithreaded network measurement ***,a cooperative communication module is designed to realize message transmission and message synchronization between the root and local controllers,and socket technology is used to ensure the reliability and stability of message transmission between multiple controllers to acquire global network state information in real ***,after the optimal intradomain and interdomain routing paths are adaptively generated by the agents in the root and local controllers,a network traffic state prediction mechanism is designed to improve awareness of the cross-domain intelligent routing method and enable the generation of the optimal routing paths in the global network in real ***–Experimental results show that the proposed cross-domain intelligent routing method can significantly improve the network throughput and reduce the network delay and packet loss rate compared to those of the Dijkstra and open shortest path first(OSPF)routing methods Originality/value–Message transmission and message synchronization for multicontroller interdomain routing in SDN have long adaptation times and slow convergence speeds,coupled with the shortcomings of traditional interdomain routing methods,such as cumbersome configuration and inflexible acquisition of network state *** drawbacks make it difficult to obtain global state information about the network,and the optimal routing decision cannot be made in real time,affecting network *** paper proposes a cross-domain intelligent SDN routing method based on a proposed MDRL ***,the network

关键词： Reinforcement learning Cross-domain intelligent routing Network traffic state prediction Software-defined network

来源：评论

学校读者我要写书评

暂无评论

Consortium Chain Consensus Vulnerability and Chain Generation Mechanism

引用

computers, Materials & Continua 2023年第11期77卷 2505-2527页

作者： Rui Qiao hi Dong School of Computer Science and Technology Zhoukou Normal UniversityZhoukou466001China

Effectively identifying and preventing the threat of Byzantine nodes to the security of distributed systems is a challenge in applying consortium ***,this paper proposes a new consortium chain generation model,deeply analyzes the vulnerability of the consortium chain consensus based on the behavior of the nodes,and points out the effects of Byzantine node proportion and node state verification on the consensus process and system ***,the normalized verification node aggregation index that represents the consensus ability of the consortium organization and the trust evaluation function of the verification node set is *** either of the two is lower than the threshold,the consortium institution or the verification node set members are dynamically *** this basis,an innovative consortium chain generation mechanism based on the Asynchronous Binary Byzantine Consensus Mechanism(ABBCM)is *** on the extended consortium chain consensus mechanism,a certain consensus value set can be combined into multiple proposals,which can realize crossdomain asynchronous message passing between multi-consortium chains without reducing the system’s *** addition,experiments are carried out under four classical Byzantine Attack(BA)behaviors,BA1 to *** results show that the proposed method can obtain better delay than the classical random Byzantine consensus algorithm Coin,effectively improving the consensus efficiency based on asynchronous message passing in the consortium chain and thus meeting the throughput of most Internet of Things(IoT)applications.

关键词： Consortium chain Byzantine consensus generation mechanism Internet of Things consensus vulnerability

来源：评论

学校读者我要写书评

暂无评论

Blockchain-based trust management for data exchange in internet of vehicle network

引用

Multimedia Tools and Applications 2025年第8期84卷 4837-4855页

作者： Srivastava, Sandeep Agarwal, Deepshikha Chaurasia, Brijesh Kumar Adhikari, Mainak Computer Science Indian Institute of Information Technology UP Lucknow226002 India Information Technology Indian Institute of Information Technology UP Lucknow226002 India Computer Science Pranveer Singh Institute of Technology UP Kanpur208029 India School of Data Science Indian Institute of Science Education & Research Thiruvananthapuram India

Internet of Vehicles (IoV) integrates with various heterogeneous nodes, such as connected vehicles, roadside units, etc., which establishes a distributed network. Vehicles are managed nodes providing all the services required during inter-vehicular communication which makes it essential to use a trust management mechanism. Existing trust management systems are insufficient to address the emerging IoV needs of transparency, mobility, immutability, and scalability. Motivated by the need to address the outlined issues, Proposed additive increase and multiplicative decrease (AIMD) trust management model, to improve the quality of communication, as well as vehicle trustworthiness based on multiple inter-vehicular interactions. The proposed layered architecture helps to reduce overhead in terms of latency and average execution time compared to other trust management mechanism and avoid risk of single point of failure. The Proposed AIMD trust management is implemented on permissioned blockchain, where transactions in the IoV network are only managed by minor nodes which accumulate interactions and uses PBFT and AIMD mechanism to derive immutable reputation score securely and transparently. Simulation results show the AIMD trust mechanism using permissioned blockchain can manage the trust of vehicles efficiently by reducing the latency by 0.29 sec, increasing packet delivery ratio from 0.8 to 0.95, and significantly reduce transaction finality time. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

GAM:A GPU-Accelerated Algorithm for MaxRS Queries in Road Networks

引用

Journal of computer science & technology 2022年第5期37卷 1005-1025页

作者： Jian Chen Kai-Qi Zhang Tian Ren Zhen-Qing Wu Hong Gao School of Computer Science and Technology Harbin Institute of TechnologyHarbin150001China

In smart phones,vehicles and wearable devices,GPS sensors are ubiquitous and collect a lot of valuable spatial data from the real *** a set of weighted points and a rectangle r in the space,a maximizing range sum(MaxRS)query is to find the position of r,so as to maximize the total weight of the points covered by r(i.e.,the range sum).It has a wide spectrum of applications in spatial crowdsourcing,facility location and traffic *** of the existing research focuses on the Euclidean space;however,in real life,the user’s moving route is constrained by the road network,and the existing MaxRS query algorithms in the road network are *** this paper,we propose a novel GPU-accelerated algorithm,namely,GAM,to tackle MaxRS queries in road networks in two phases *** phase 1,we partition the entire road network into many small cells by a grid and theoretically prove the correctness of parallel query results by grid shifting,and then we propose an effective multi-grained pruning technique,by which the majority of cells can be pruned without further *** phase 2,we design a GPU-friendly storage structure,cell-based road network(CRN),and a two-level parallel framework to compute the final result in the remaining ***,we conduct extensive experiments on two real-world road networks,and the experimental results demonstrate that GAM is on average one order faster than state-of-the-art competitors,and the maximum speedup can achieve about 55 times.

关键词： road network maximizing range sum GPU acceleration pruning strategy

来源：评论

学校读者我要写书评

暂无评论

Enhanced secure medical data sharing with traceable and direct revocation

引用

The Journal of China Universities of Posts and Telecommunications 2023年第1期30卷 66-79页

作者： Peng Weiping Cui Shuang Song Cheng Han Ning School of Computer Science and Technology Henan Polytechnic UniversityJiaozuo 454003China

Sharing of the electronic medical records among different hospitals raises serious concern of the leakage of individual privacy for the adoption of the semi trustworthiness of the medical cloud *** tracking and revocation of malicious users have become urgent *** solve these problems,this paper proposed a traceable and directly revocable medical data sharing *** the scheme,a unique identity parameter(ID),which was generated and embedded in the private key generation phase by the medical service provider(MSP),is used to identify legal authorized user and trace malicious *** when attributes satisfy the access policy and user’s ID is not in the revocation list can the user calculate the decryption *** user can be tracked and directly revoked by using the revocation *** the assumption of decision bilinear Diffie-Hellman(DBDH),this paper has proved that the scheme is able to achieve security against chosen-plaintext attack(CPA).The performance analysis demonstrates that the sizes of the public key and private key are shorter,and the time overhead is less than other schemes in the public-private key generation,data encryption and data decryption stages.

关键词： medical data sharing electronic medical records tracking and revocation privacy protection

来源：评论

学校读者我要写书评

暂无评论

Probabilistic fault diagnosis of clustered faults for multiprocessor systems

引用

Journal of computer science & technology 2023年第4期38卷 821-833页

作者：孙雪丽樊建席程宝雷王岩张力 School of Computer Science and Technology Soochow UniversitySuzhou 215006China Provincial Key Laboratory for Computer Information Processing Technology Soochow UniversitySuzhou 215006China

With the development of high-performance computing and the expansion of large-scale multiprocessor sys-tems,it is significant to study the reliability of *** fault diagnosis is of practical value to the reliabilityanalysis of multiprocessor *** this paper,we design a linear time diagnosis algorithm with the multiprocessor sys-tem whose threshold is set to 3,where the probability that any node is correctly diagnosed in the discrete state can be ***,we give the probabilities that all nodes of a d-regular and d-connected graph can be correctly diag-nosed in the continuous state under the Weibull fault distribution and the Chi-square fault *** prove thatthey approach to 1,which implies that our diagnosis algorithm can correctly diagnose almost all nodes of the graph.

关键词： regular graph clustered fault probabilistic diagnosis algorithm Preparata Metze Chien model(PMCmodel)

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：