检索结果-内蒙古大学图书馆

5th Asia-Pacific Conference on Image Processing, Electronics and computers, IPEC 2024

作者： Yang, Chen Gu, Zheyuan Bai, Jing Li, Zhe Xiong, Gang Gou, Gaopeng Yao, Shan Chen, Xi Institute of Information Engineering Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences School of Cyber Security Beijing China Zhengzhou University School of Computer and Artificial Intelligence Zhengzhou China 360 Digital Security Technology Group Co. Ltd. Beijing China

ISBN: (纸本)9798350374407

Encrypted traffic classification is an essential technique for maintaining cybersecurity and assuring quality of experience. Current research has achieved excellent classification performance by utilizing machine learning or deep learning methods. However, the constantly emerging new encryption services make collecting labeled traffic samples increasingly difficult, failing to meet the training requirements of pre-existing models. Therefore, encrypted traffic classification with few-shot learning has emerged as an increasingly popular research area. In this paper, we conduct a comprehensive survey and summary of the latest advancements in few-shot encrypted traffic classification, categorizing existing methods into: data augmentation-based, transfer learningbased, metric learning-based, meta-learning-based, and pretraining-based approaches. We introduce the principles and typical works of these methods and discuss their advantages and limitations. Finally, we summarize the research findings and provided insight into potential avenues for future inquiry. © 2024 IEEE.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

A Novel Approach for Detecting and Preventing security attacks using Machine Learning in IoT 14

A Novel Approach for Detecting and Preventing Security attac...

引用

14th International Conference on Computing Communication and Networking Technologies, ICCCNT 2023

作者： Al Sukhni, Badeea Kumar Mohanta, Bhabendu Kumar Dehury, Mohan Kumar Tripathy, Asesh University of Westminster Department Cyber Security London United Kingdom Koneru Lakshmaiah Education Foundation Department of Computer Science and Engineering Andhra Pradesh Vaddeswaram India Amity University Jharkhand Ranchi Amity Inst. of Information Technology India

ISBN: (纸本)9798350335095

Since most of the applications rely on machine learning algorithms, the customers are affected by the machine-made decisions. This can be resolved by adding more complexity to ML models. Common attacks that are possible to break the reliability of ML techniques are replay attacks, zero-dynamics attacks, and adversarial attacks. These attacks leverage the vulnerabilities of ML and collapse the reliability of data. The development of trusted ML systems can be achieved with the help of technical tools such as probabilistic modeling, Markov decision process. temporal decision process. Then, to achieve a reliable machine-learning framework, a simple way could be to utilize an ensemble for the regular models. This can develop reliable predictions. Another emerging technique is transfer learning that helps to overcome the performance loss and increases the trust score. However, estimating reliability will define the accuracy of predictions, it can work along with transductive reasoning. In general, reliability calculation is defined as the probability of count of true classes from the example set. © 2023 IEEE.

关键词： Reliability

来源：评论

学校读者我要写书评

暂无评论

Optimal, Almost Optimal Few-Weight Linear Codes and Related Quantum Codes

引用

IEEE Transactions on Information Theory 2025年第6期71卷 4250-4259页

作者： Xie, Conghui Chen, Hao Zhou, Haiyan Li, Yang Lao, Huimin Jinan University College of Information Science and Technology Cyber Security Guangdong Guangzhou510632 China Nanjing Normal University Institute of Mathematics School of Mathematical Sciences Jiangsu Nanjing210023 China Hefei University of Technology School of Mathematics Anhui Hefei230009 China National University of Singapore Department of Industrial Systems Engineering and Management Queenstown 119077 Singapore

In eight published papers in IEEE Transactions on Information Theory, infinite families of optimal few-weight binary and q-ary linear codes were constructed and their weight distributions were determined. These codes are linear codes meeting the Griesmer bound. We indicate that many Griesmer codes constructed in these papers are not new. They are actually Solomon-Stiffler codes invented in 1965. Therefore weight distributions of some special binary or q-ary Solomon-Stiffler codes were determined in the papers mentioned above. From a similar geometric approach as Solomon-Stiffler codes, we construct ten infinite families of binary, ternary and quaternary few-weight, optimal, almost optimal and near-optimal linear codes close to the Griesmer bound and their weight distributions are determined. These linear codes have positive Griesmer defects up to five, and thus not Solomon-Stiffler codes and Griesmer codes from minihypers. Moreover, many optimal, best known and almost optimal quantum codes of small lengths, comparing with Grassl’s table on quantum codes, are constructed from the same geometric approach as binary Solomon-Stiffler codes. © 1963-2012 IEEE.

关键词： Binary codes

来源：评论

学校读者我要写书评

暂无评论

Intersecting-Boundary-Sensitive Fingerprinting for Tampering Detection of DNN Models 41

Intersecting-Boundary-Sensitive Fingerprinting for Tampering...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Bai, Xiaofan He, Chaoxiang Ma, Xiaojing Zhu, Bin Benjamin Jin, Hai School of Cyber Science and Engineering Huazhong University of Science and Technology China National Engineering Research Center for Big Data Technology and System China Services Computing Technology and System Lab China Hubei Engineering Research Center on Big Data Security China Hubei Key Laboratory of Distributed System Security China Microsoft United States School of Computer Science and Technology Huazhong University of Science and Technology China Cluster and Grid Computing Lab China

Cloud-based AI services offer numerous benefits but also introduce vulnerabilities, allowing for tampering with deployed DNN models, ranging from injecting malicious behaviors to reducing computing resources. Fingerprint samples are generated to query models to detect such tampering. In this paper, we present Intersecting-Boundary-Sensitive Fingerprinting (IBSF), a novel method for black-box integrity verification of DNN models using only top-1 labels. Recognizing that tampering with a model alters its decision boundary, IBSF crafts fingerprint samples from normal samples by maximizing the partial Shannon entropy of a selected subset of categories to position the fingerprint samples near decision boundaries where the categories in the subset intersect. These fingerprint samples are almost indistinguishable from their source samples. We theoretically establish and confirm experimentally that these fingerprint samples' expected sensitivity to tampering increases with the cardinality of the subset. Extensive evaluation demonstrates that IBSF surpasses existing state-of-the-art fingerprinting methods, particularly with larger subset cardinality, establishing its state-of-the-art performance in black-box tampering detection using only top-1 labels. The IBSF code is available at: https://***/CGCL-codes/IBSF. Copyright 2024 by the author(s)

关键词： HTTP

来源：评论

学校读者我要写书评

暂无评论

Three-dimensional dynamic uncertainty semantic SLAM method for a production workshop

Three-dimensional dynamic uncertainty semantic SLAM method f...

引用

作者： Gou, Rongsong Chen, Guangzhu Yan, Chengliang Pu, Xin Wu, Yuanyuan Tang, Yuan College of Mechanical and Electrical Engineering Chengdu University of Technology Chengdu China College of Computer Science and Cyber Security Chengdu University of Technology Chengdu China

Simultaneous localization and mapping (SLAM) is the basis for intelligent navigation and intelligent security of mobile robots in a workshop. However, there are numerous dynamic objects in a workshop, such as robots and operators, which can decrease the accuracy of robot localization and mapping. To solve this problem, this paper proposes a three-dimensional dynamic semantic system named the PW_SLAM. First, the production workshop-oriented lightweight semantic segmentation network (PWnet) is integrated with the SLAM system by a robot operating system(ROS) to provide semantic information. Next, a dynamic uncertainty keypoint classifier(DUKC) is proposed to filter the dynamic keypoints and improve the localization accuracy. Then, the map is stored in the form of an octree, and a dynamic object map filter are designed to filter the dynamic objects in the octree map. The proposed system is evaluated on the TUM RGB-D public datasets and the results show that the RMSE of absolute trajectory error in PW_SLAM is decreased by 92.80%, 98.85%, 87.21% compared to that of ORB_SLAM3 on the TUM highly dynamic datasets "fr3/walking_xyz", "fr3/walking_static", and "fr3/walking_rpy", respectively. In addition, the PW_SLAM is also applied to the real workshop environment, and the results show that it can effectively eliminate dynamic objects and build a static octree map of a workshop environment effectively. © 2022 Elsevier Ltd

关键词： Robot Operating System

来源：评论

学校读者我要写书评

暂无评论

A Research on the Academic System in Universities Based on the One-Hot Encoding PAC Fuzzy Comprehensive Evaluation Algorithm 7th

A Research on the Academic System in Universities Based on t...

引用

7th International Conference on Innovative Computing, IC 2024

作者： Lu, Zhiqian Liu, Yihan Li, Qiang College of Computer and Cyber Security Hebei Normal University Hebei Shijiazhuang China College of China Combustion Engineering Hebei Normal University Hebei Shijiazhuang China College of Career Technology Hebei Normal University Hebei Shijiazhuang China

ISBN: (纸本)9789819741922

The influence of contemporary learning exhibits complex and multifaceted characteristics. This study, building upon a detailed explanation of the model, constructs a comprehensive set of evaluation indicators. By selecting survey questionnaire data from the 15th Electrical engineering Cup on "Evaluation of the Impact of Artificial Intelligence on College Students’ Learning" as a sample, this research proposes specific steps for a study on the impact of artificial intelligence on college students’ learning based on the fuzzy comprehensive evaluation method. Initially, the one-hot encoding method ‘get dummies’ in pandas is used to process non-continuous numerical features, thereby fuzzifying the data. Subsequently, Principal Component Analysis (PCA) is employed to centralize the data, reducing its dimensionality by calculating covariance, eigenvalues, and eigenvectors. The primary components are then selected, and data is projected onto these components to obtain a dimensionally reduced dataset. The main method used in this article is PCA algorithm implementation through SVD decomposition of the covariance matrix for the purpose of data classification. The research results reveal that the fuzzy comprehensive evaluation model not only integrates subjective and objective factors but also allows for qualitative adjustments and batch processing of data, making the evaluation results more comprehensive and accurate. This effectively aids in a deeper understanding of the impact of artificial intelligence on college students’ learning and provides robust support for related decision-making outcomes. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

MICABAC: Multidimensional Industrial Control Attribute-Based Access Control Model

MICABAC: Multidimensional Industrial Control Attribute-Based...

引用

2024 IEEE International Conference on Systems, Man, and cybernetics, SMC 2024

作者： Wang, Hangyu Lv, Fei Chen, Yuqi Si, Shuaizong Pan, Zhiwen Sun, Degang Sun, Limin Institute of Information Engineering Cas Beijing Key Laboratory of Iot Information Security Technology Beijing China School of Cyber Security University of Chinese Academy of Sciences Beijing China Shanghai Tech University Shanghai China Computer Network Information Center Chinese Academy of Sciences Beijing China

ISBN: (纸本)9781665410205

As the Industrial Control System (ICS) increasingly merges with the Internet, the security threats have been increasing from internal users and external hackers. These challenges are further intensified by the facts: industrial control devices and protocols, leading to the inadequacy of traditional access control models in tackling the intricacies of ICS. We identify attributes that are optimally aligned with the specific needs of the ICS environment and propose the Multidimensional Industrial Control Attribute-Based Access Control Model (MICABAC) as a customized solution. MICABAC model significantly improves access control security and is finer granularity by selecting and evaluating required attributes within various ICS. We have been validated in two real-world ICS environments: the Gas Pipe Network System (GPNS) and the computer Numerical Control (CNC) machine tool. Experiments indicate that by integrating MICABAC into the existing system, the maximum delay for access requests is 63.83 ms. In terms of accuracy in defending against malicious attacks, the GPNS achieves 96.49% and the CNC reaches 94.86%. Finally, we discuss the advantages and limitations of MICABAC and explore potential directions for future research. © 2024 IEEE.

关键词： ABAC Access Control Industrial Control System PLC

来源：评论

学校读者我要写书评

暂无评论

Block-based fine-grained and publicly verifiable data deletion for cloud storage: Block-based fine-grained and publicly verifiable..: C. Yang et al.

引用

Soft Computing 2024年第21期28卷 12491-12506页

作者： Yang, Changsong Liu, Yueling Ding, Yong Wu, Yongqiang Guangxi Key Laboratory of Cryptography and Information Security Guilin University of Electronic Technology Guilin541004 China Lion Rock Labs of Cyberspace Security Institute of Cyberspace Technology HKCT Institute for Higher Education 999077 Hong Kong Guangxi Engineering Research Center of Industrial Internet Security and Blockchain Guilin University of Electronic Technology Guilin541004 China School of Computer Guangdong University of Science & Technology Dongguan523083 China South China Technical Support Department Topsec Network Technology Inc. Guangzhou510000 China

One of the most important services provided by cloud service providers (CSPs), cloud storage is economically attractive and can provide on-demand data storage service to resource-constrained tenants in the manner of pay-per-use. Therefore, by embracing cloud storage service, tenants can store their large-scale data on the remote cloud host to tremendously reduce local storage burden and computational overhead. However, due to the separation of cloud data ownership and management, cloud storage inevitably suffers from a few new security challenges, such as fine-grained cloud data deletion. To achieve data deletion in cloud storage, we design a vector-commitment-based publicly verifiable data deletion scheme, which can achieve data integrity auditing and block-based fine-grained data deletion. Specifically, our proposed scheme disguises the data deletion operation as data update operation. Therefore, the tenants can achieve data deletion by updating the cloud data blocks with some unrelated data blocks. Meanwhile, the tenants can prevent cloud data from being polluted by periodically performing data integrity auditing operation, thus guaranteeing that the cloud host sincerely maintains the data. Moreover, we formally analyze the security, which shows that our proposed scheme can satisfy all of the anticipative security requirements without interaction with a centralized third party. Finally, we develop a prototype system and evaluate the performance of our proposed scheme, which can intuitively demonstrate the high-efficiency and practicality of our proposal. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024.

关键词： Data integrity

来源：评论

学校读者我要写书评

暂无评论

An Improved Multiple Face Recognition System for Crowd Monitoring Applications Based on Transfer Learning Approach

An Improved Multiple Face Recognition System for Crowd Monit...

引用

International Conference on 6G Communications Networking and Signal Processing, SGCNSP 2023

作者： Jayasree, M. Sunitha, K.A. Brindha, A. Punna, Rajasekhar Aravamuthan, G. Retnakumar, G. Joselin Department of Electronics and Instrumentation Engineering College of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Tamilnadu Chennai603203 India Department of Electronics and Communication Engineering School of Engineering and Sciences SRM University Mangalagiri Andhra Pradesh Amaravati522502 India Security Electronics and Cyber Technology Division Bhabha Atomic Research Centre Anushakti Nagar Maharashtra Mumbai400085 India

ISBN: (纸本)9789819778751

The real-time Face Recognition (FR) techniques are limited to situations where only one face is visible in close proximity to the camera. Sometimes, the people under surveillance may not be directly facing the camera while walking. However, there is a growing need for FR technology to accurately identify multiple faces in crowded areas even when individuals are not facing the camera. An AI-based multiple-face recognition (MFR) system has been developed to improve performance by incorporating an increased number of pose variation samples. The developed system utilizes a pre-trained FaceNet architecture, which converts the face images into a compact Euclidean space of dimension 128 × 1. This study focuses on improving accuracy and decreasing computation time for multiple faces within the field of view. Results show that the FaceNet model has a high recognition rate of 99.7%. The system can recognize up to 10 faces in the field of view at a computational time of 1.21 s. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

An NTT/INTT Accelerator with Ultra-High Throughput and Area Efficiency for FHE 24

An NTT/INTT Accelerator with Ultra-High Throughput and Area ...

引用

61st ACM/IEEE Design Automation Conference, DAC 2024

作者： Lu, Zhaojun Yu, Weizong Xu, Peng Wang, Wei Zhang, Jiliang Feng, Dengguo Hubei Key Laboratory of Distributed System Security School of Cyber Science and Engineering Hust Wuhan JinYinHu Laboratory Hubei Wuhan China School of Cyber Science and Engineering Hust Hubei Wuhan China School of Computer Science and Technology Hust Hubei Wuhan China Hnu Hunan Changsha China State Key Laboratory of Cryptology Beijing China

ISBN: (纸本)9798400706011

As a core arithmetic operation and security guarantee of Fully Homomorphic Encryption (FHE), Number Theoretic Transform (NTT) of a large degree is the primary source of computational and time overhead. In this paper, we propose a scalable and conflict-free memory mapping algorithm that breaks the memory bound and releases a large amount of on-chip resources. A flexible and no-stall hardware/software pipeline architecture is designed to boost the throughput of NTT/INTT of N = 216 to over 48,543 operations per second with area efficiency, which 4× and 10× speed up the FPGA-based (HPCA'23) and GPU-based (HPCA'23) schemes. © 2024 Copyright is held by the owner/author(s). Publication rights licensed to ACM.

关键词： Memory architecture

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：