检索结果-内蒙古大学图书馆

2023 IEEE International Conference on Dependable, Autonomic and Secure Computing, 2023 International Conference on Pervasive Intelligence and Computing, 2023 International Conference on Cloud and big data Computing, 2023 International Conference on Cyber science and technology Congress, DASC/PiCom/CBDCom/CyberSciTech 2023

作者： Xiong, Maolin Chen, Yuling Dou, Hui Luo, Yun Tan, Chaoyue Zhang, Yangwen College of Computer Science and Technology Guizhou University State Key Laboratory of Public Big Data Guiyang China

ISBN: (纸本)9798350304602

Deep neural network is vulnerable to adversarial samples, which means altering the image subtly to introduce errors into the neural network model. Adversarial samples can categorize into black-box and white-box attacks, white-box attacks have been able to achieve a high success rate, but the performance of black-box attacks is weak due to the large gap between the substitute model and the victim model. Previously, the way to enhance the substitute model is image transformations on the spatial domain, to simulate different victim models. However, these methods lead to weak transferability, because it can not generate diverse enhanced models. To solve this problem, we propose a dual-frequency-domain transform attack based on DCT(Discrete Cosine Transform) and DWT(Discrete Wavelet Transformation), proposed method optimizes the process of frequency-domain transform, which can greatly enhance the substitute models. Besides, our method can combine with others (gradient attack methods and spatial domain attack methods) to further enhance the attack effect. Experiments show that the attack success rate of our method is good in several mainstream no-defense models. © 2023 IEEE.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

A Blockchain of Blockchains Structure Based on Asset Trading Scheme

A Blockchain of Blockchains Structure Based on Asset Trading...

引用

作者： Huang, Siyuan Chen, Yuling Tan, Chaoyue Dou, Hui Luo, Yun Yang, Yuxiang College of Computer Science and Technology Guizhou University State Key Laboratory of Public Big Data Guiyang China

ISBN: (纸本)9798350304602

In the realm of blockchain applications, the issue of value isolation, commonly referred to as the value island problem, imposes inherent limitations on the applicability of corresponding assets. To overcome this challenge, researchers have dedicated their efforts to exploring diverse solutions for facilitating cross-chain transactions involving various assets. These solutions include centralized exchanges and hash time lock protocols. However, these approaches are subject to significant limitations. Centralized exchanges, while providing transactional functionalities, undermine the decentralized nature intrinsic to blockchain technology. Conversely, hash time lock protocols necessitate a high level of trust between transacting parties. Furthermore, although decentralized exchanges have emerged to support atomic cross-chain exchanges, their transactional architectures are restricted to either a bilateral relationship between two blockchains or a subset of pre-established blockchains. Therefore, the construction of a non-trust-based, decentralized exchange that supports multiple blockchain networks remains an ongoing challenge. In this research, we propose Vchain, a cross-chain relay transaction scheme, designed with a BoBs(Blockchain of Blockchains) structure, to address the aforementioned challenges. Vchain enables the participation of various accounts within blockchain ecosystems that support smart contract functionalities. It ensures the security and integrity of user transactions by leveraging third-party cryptocurrency assets as collateral. Moreover, users have the flexibility to selectively choose transaction counterparties based on the quantity of collateral assets provided, thereby maximizing their transactional interests. Consequently, Vchain enables non-trust-based, multi-chain asset cross-chain transactions, addressing the limitations of existing solutions. © 2023 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Robust Object Detection Based on a Comparative Learning Perspective

Robust Object Detection Based on a Comparative Learning Pers...

引用

作者： Yang, Hao Chen, Yuling Dou, Hui Luo, Yun Tan, Chaoyue Zhang, Yangwen College of Computer Science and Technology Guizhou University State Key Laboratory of Public Big Data Guiyang China

ISBN: (纸本)9798350304602

With the advent of adversarial samples, the safety of Artificial Intelligence is of particular concern, as only a small amount of perturbation can be added to mislead a model's judgment. Therefore there is an urgent need for research on models that can resist adversarial perturbations. To alleviate this problem, we first analyze the vulnerability of adversarial samples and propose a uniform perspective robust model for object detection that accurately identifies both clean and ad-versarial samples. We propose a robust object detection based on the contrastive learning perspective (RCP), which can learn features from both clean and adversarial samples from a fine-grained perspective, and can recognize adversarial samples more accurately. Extensive experiments on PASCAL VOC and MS COCO show that our proposed method only degrades the clean sample detection performance by a small amount in exchange for a large robustness improvement against adversarial attacks, achieving state of the art results. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

A High Capacity Combined Carrier Steganography Without Embedding Image

A High Capacity Combined Carrier Steganography Without Embed...

引用

作者： Zhang, Yangwen Chen, Yuling Dou, Hui Luo, Yun Tan, Chaoyue Xiong, Maolin College of Computer Science and Technology Guizhou University State Key Laboratory of Public Big Data Guiyang China

ISBN: (纸本)9798350304602

The technique of image steganography is the embedding of secret information into the carrier image by means of algorithms that are widely used in the field of information transmission, but steganalysis tools built with traditional steganographic algorithms can easily identify them. The technique of steganography without embedding (SWE) is the mapping of the secret information onto the noise and the generation of an image by mapping the noise of the secret information. This SWE technique is difficult to find without directly using secret information. But the capacity of secret information is less and the extraction accuracy is low when the SWE is currently hidden. Therefore, a steganography without embedding technique for carrier secret information is proposed in this paper based on the above problems. Initially, the secret information is mapped into the carrier noise, which can increase the secret information capacity of steganography and improve the accuracy of extracting information. Finally, the effectiveness and generality of the design method will be verified by experiments. © 2023 IEEE.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

A Multiple Indicator Credit Ranking Scheme Based on Notary Mechanism

A Multiple Indicator Credit Ranking Scheme Based on Notary M...

引用

作者： Chen, Lang Chen, Yuling Tan, Chaoyue Dou, Hui Luo, Yun Chen, Pengyu College of Computer Science and Technology Guizhou University State Key Laboratory of Public Big Data Guiyang China

ISBN: (纸本)9798350304602

Blockchain has penetrated multiple fields such as finance, healthcare, supply chain, and intelligent transportation. However, the value exchange between different independent blockchain systems limits the expansion of blockchain. Cross-chain technology can achieve the transfer of digital assets between blockchain networks, and the notary mechanism is a typical cross-chain technology that can achieve value exchange between different blockchains. However, it requires trust in specific notaries. To address these issues, this paper proposes a multi-indicator credit ranking scheme based on notaries to improve the reliability of notaries and ensure the security of value exchange between different blockchains. To tackle the problem of single and incomplete inherent value evaluation indicators in traditional notary mechanisms, a multi-indicator evaluation is adopted to ensure the fairness of notary node credit evaluation. To address the issue of inaccurate comprehensive credit evaluation of notaries, a comprehensive evaluation and ranking of notary nodes are conducted based on their historical transaction information to improve the accuracy of their credit evaluation. The experimental results show that using multiple indicators to evaluate the inherent value of notary nodes can comprehensively evaluate the scheme and accurately screen out high-credit nodes to act as notaries, improving the security of cross-chain asset transactions. © 2023 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Communication-Efficient Federated Learning with Sparsity and Quantization

Communication-Efficient Federated Learning with Sparsity and...

引用

作者： Long, Zhong Chen, Yuling Dou, Hui Luo, Yun Tan, Chaoyue Sun, Yancheng College of Computer Science and Technology Guizhou University State Key Laboratory of Public Big Data Guiyang China

ISBN: (纸本)9798350304602

Google proposed federated learning in 2016, which aims to solve the problems of data silos and privacy leakage in machine learning. However, as the difficulty of the target task increases and the model performance requirements progressively improve, these have to consume more communication costs to increase the model parameters. In order to solve this problem, this paper proposes the Federated Learning with Sparsity and Quantization(FedSQ) method, where we select the gradients with a large rate of change in the gradients of the participants in each round for transmission, and utilize binary compression to quantize the selected model parameters, while adopting a suitable aggregation algorithm to replace the original FedAvg, so that it does not excessively impair the model accuracy in the case of high-fold compression. Finally, we verify the effectiveness of the scheme proposed in this paper, which can improve the quality of model training under model compression. © 2023 IEEE.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Optimization of cloud computing task scheduling based on whale optimization algorithm with quasi-opposition-based and nonlinear factor 6

Optimization of cloud computing task scheduling based on wha...

引用

6th International Conference on Internet of Things, Automation and Artificial Intelligence, IoTAAI 2024

作者： Xuan, Xiong Li, Shaobo Yu, Liya Wu, Fengbin Zhang, Xingxing An, Xue College of Computer Science and Technology Guizhou University State Key Laboratory of Public Big Data Guiyang550025 China Guizhou University State Key Laboratory of Public Big Data Guiyang550025 China Guizhou University College of Mechanical Engineering Guiyang550025 China

ISBN: (纸本)9798350386974

Task scheduling has a significant impact on the resource availability and operation cost of the system in cloud computing. In order to improve the efficiency of task execution in the cloud for a given computational resource, the improved whale optimization algorithm is applied to the multi-objective optimization model for cloud task scheduling. Specifically, the quasi-opposition-based learning strategy is added in the initiation phase to increase the diversity of the population. Meanwhile, a novel nonlinear convergence factor was introduced to augment the local search capability. Experimental results indicate that the improved whale optimization algorithm converges faster and searches with higher precision in finding the optimal task scheduling solution compared to existing meta-heuristic algorithms. In addition, the modified algorithm achieves superior system resource usage when dealing with both small and large tasks. © 2024 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

MMH-FE:AMulti-Precision and Multi-Sourced Heterogeneous Privacy-Preserving Neural Network Training Based on Functional Encryption

引用

computers, Materials & Continua 2025年第3期82卷 5387-5405页

作者： Hao Li Kuan Shao Xin Wang Mufeng Wang Zhenyong Zhang The State Key Laboratory of Public Big Data College of Computer Science and TechnologyGuizhou UniversityGuiyang550025China Key Laboratory of Computing Power Network and Information Security Ministry of EducationShandong Computer Science CenterQilu University of Technology(Shandong Academy of Sciences)Jinan250014China China Industrial Control Systems Cyber Emergency Response Team Beijing100040China

Due to the development of cloud computing and machine learning,users can upload their data to the cloud for machine learning model ***,dishonest clouds may infer user data,resulting in user data *** schemes have achieved secure outsourced computing,but they suffer from low computational accuracy,difficult-to-handle heterogeneous distribution of data from multiple sources,and high computational cost,which result in extremely poor user experience and expensive cloud computing *** address the above problems,we propose amulti-precision,multi-sourced,andmulti-key outsourcing neural network training ***,we design a multi-precision functional encryption computation based on Euclidean ***,we design the outsourcing model training algorithm based on a multi-precision functional encryption with multi-sourced ***,we conduct experiments on three *** results indicate that our framework achieves an accuracy improvement of 6%to 30%.Additionally,it offers a memory space optimization of 1.0×2^(24) times compared to the previous best approach.

关键词： Functional encryption multi-sourced heterogeneous data privacy preservation neural networks

来源：评论

学校读者我要写书评

暂无评论

Face Recognition Based on SRCS Algorithm and Score of Exponential Weighting

Face Recognition Based on SRCS Algorithm and Score of Expone...

引用

International Conference on Image, Vision and Intelligent Systems, ICIVIS 2023

作者： Zhang, Xuexue Zhang, Yongjun Gao, Weihao Long, Wei Yao, He Sate Key Laboratory of Public Big Data College of Computer Science and Technology Guizhou University Guizhou Guiyang550025 China

ISBN: (纸本)9789819708543

Face recognition is a prominent research area within computer vision, offering significant value across various fields of life, including certificate verification, security systems, and human-computer interactions. Nonetheless, enhancing the reliability and precision of face recognition systems is persistently challenging due to variables like facial expressions, various scenes, and occlusion. This paper introduces an enhanced SRCS (Sum of Representation Coefficients as Score) algorithm, which builds upon the two-stage sparse representation approach. The underlying principle is that a higher representation value of each class of training images vis-à-vis test images suggests a stronger correlation between the test image and the respective class. Consequently, the algorithm computes the sum of representation coefficients for the training images to ascertain the distances between the classes and the test images. Following this, it identifies classes with high scores as prospective representatives of the test samples. Moreover, the setting of reasonable weights in the weighted fusion of addition weighting, based on the TSFR (The Two-Step Face Recognition) method, presents challenges, and varying training samples require different optimal weights. Therefore, this paper proposes an exponential weighting method score that obviates the need for setting any weights while effectively enhancing recognition accuracy. This proposed method undergoes rigorous validation through comprehensive experiments conducted on renowned databases such as FERET, PRL, YALE, and GT, with the results consistently affirming the effectiveness of the method. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

An enhanced exponential distribution optimizer for solving task scheduling in cloud computing systems 4

An enhanced exponential distribution optimizer for solving t...

引用

4th International Conference on Advanced Algorithms and Neural Networks, AANN 2024

作者： Wu, Fengbin Li, Shaobo Wu, Libang Xiong, Xuan Xie, Rongxiang State Key Laboratory of Public Big Data Guizhou University China Guizhou Institute of Technology China College of Computer Science and Technology Guizhou University China

ISBN: (纸本)9781510686106

Cloud computing has become integral to information technology, offering a flexible and scalable way to access and utilize computing resources. In cloud computing, task scheduling policies can affect the resource usage efficiency of the underlying system. Hence, allocating user-input tasks to appropriate computing resources is an essential issue in cloud task scheduling. For this, many meta-heuristic algorithms have been introduced. This paper addresses the problem of cloud task scheduling using the exponential distribution optimizer (EDO) for the first time. Meanwhile, an enhanced EDO variant, called EEDO, is proposed to enhance the optimal solution search ability further. Specifically, a chaotic oppositional learning strategy is proposed to improve population diversity. Then, a guided solution weight adaptation technique is developed to augment the accuracy. Experimental findings indicate that the proposed EEDO makes better use of system resources for different tasks. Hence, EEDO can be considered as a promising scheduling method. © 2024 SPIE.

关键词： Scheduling algorithms

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：