检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Lai, Yingxin Yu, Zitong Yang, Jing Li, Bin Kang, Xiangui Shen, Linlin The School of Computing and Information Technology Great Bay University Dongguan523000 China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen518060 China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen University Shenzhen518060 China The Guangdong Key Laboratory of Information Security The School of Computer Science and Engineering Sun Yat-sen University Guangzhou510080 China Computer Vision Institute School of Computer Science & Software Engineering Shenzhen Institute of Artificial Intelligence and Robotics for Society Guangdong Key Laboratory of Intelligent Information Processing National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen518060 China

Existing face forgery detection usually follows the paradigm of training models in a single domain, which leads to limited generalization capacity when unseen scenarios and unknown attacks occur. In this paper, we elaborately investigate the generalization capacity of deepfake detection models when jointly trained on multiple face forgery detection datasets. We first find a rapid degradation of detection accuracy when models are directly trained on combined datasets due to the discrepancy across collection scenarios and generation methods. To address the above issue, a Generalized Multi-Scenario Deepfake Detection framework (GM-DF) is proposed to serve multiple real-world scenarios by a unified model. First, we propose a hybrid expert modeling approach for domain-specific real/forgery feature extraction. Besides, as for the commonality representation, we use CLIP to extract the common features for better aligning visual and textual features across domains. Meanwhile, we introduce a masked image reconstruction mechanism to force models to capture rich forged details. Finally, we supervise the models via a domain-aware meta-learning strategy to further enhance their generalization capacities. Specifically, we design a novel domain alignment loss to strongly align the distributions of the meta-test domains and meta-train domains. Thus, the updated models are able to represent both specific and common real/forgery features across multiple datasets. In consideration of the lack of study of multi-dataset training, we establish a new benchmark leveraging multi-source data to fairly evaluate the models' generalization capacity on unseen scenarios. Both qualitative and quantitative experiments on five datasets conducted on traditional protocols as well as the proposed benchmark demonstrate the effectiveness of our approach. The codes will be available on https://***/laiyingxin2/GM-DF. Copyright © 2024, The Authors. All rights reserved.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

A Continuous Verification Mechanism for Clients in Federated Unlearning to Defend the Right to be Forgotten

A Continuous Verification Mechanism for Clients in Federated...

引用

International Symposium on Parallel and Distributed Processing with Applications, ISPA

作者： Shanshan Chen Jun Liu Yuying Li Yamin Li Yang Yang Gaoyang Liu Chen Wang Key Laboratory of Intelligent Sensing System and Security (Ministry of Education) School of Artificial Intelligence Hubei University Wuhan China School of Computer Science and Information Engineering Hubei University Wuhan China Huazhong University of Science and Technology Wuhan China

ISBN: (数字)9798331509712

ISBN: (纸本)9798331509729

In Federated Learning (FL), the regulatory need for the "right to be forgotten" requires efficient Federated Unlearning (FU) methods, which enable FL models to unlearn appointed training data. Associating with the emergence FU, verifying the performance of FU plays a critical role in evaluating the consistency FU methods, in case of the unexpected degradation of the FL model. Though well developed, none of the existing verification methods in FU stands for the clients who opt out of the FL process, which is a universal demand in FL. More specifically, after the clients quit the FL cooperation, they can no longer verify whether the FL model unlearns their data after the FL keeps training for several rounds. To this end, we introduce a continuous verification mechanism for FL clients, called Backdoor Attack-based Forgetting Verification (BAFV). Inspired by backdoor attack, BAFV embedded a persistent mark for the client that proposes to leave, with the intention that the client still has the right to verify of FU after leaving the FL cooperation for a relatively long period. Extensive experiments across diverse FU environments and datasets demonstrated that our method maintains the accuracy of model and provides clients with a continuous verification mechanism to defend their rights. Our code of BAFV is publicly available at: https://***/paper-liu/***.

关键词： Training Privacy Distributed processing Accuracy Federated learning Training data Feature extraction Data models Servers Protection

来源：评论

学校读者我要写书评

暂无评论

Scavenger: Better Space-Time Trade-Offs for Key-Value Separated LSM-trees

Scavenger: Better Space-Time Trade-Offs for Key-Value Separa...

引用

International Conference on Data engineering

作者： Jianshun Zhang Fang Wang Sheng Qiu Yi Wang Jiaxin Ou Junxun Huang Baoquan Li Peng Fang Dan Feng Wuhan National Laboratory for Optoelectronics Key Laboratory of Information Storage System Engineering Research Center of data storage systems and Technology Ministry of Education of China School of Computer Science and Technology Huazhong University of Science and Technology China Shenzhen Huazhong University of Science and Technology Research Institute Bytefrance Inc.

ISBN: (数字)9798350317152

ISBN: (纸本)9798350317169

Key- Value Stores (KVS) implemented with log- structured merge-tree (LSM-tree) have gained widespread ac-ceptance in storage systems. Nonetheless, a significant challenge arises in the form of high write amplification due to the compaction process. While KV-separated LSM-trees successfully tackle this issue, they also bring about substantial space am-plification problems, a concern that cannot be overlooked in cost-sensitive scenarios. Garbage collection (GC) holds significant promise for space amplification reduction, yet existing GC strategies often fall short in optimization performance, lacking thorough consideration of workload characteristics. Additionally, current KV-separated LSM-trees also ignore the adverse effect of the space amplification in the index LSM-tree. In this paper, we systematically analyze the sources of space amplification of KV- separated LSM-trees and introduce Scavenger, which achieves a better trade-off between performance and space amplification. Scavenger initially proposes an I/O-efficient garbage collection scheme to reduce I/O overhead and incorporates a space-aware compaction strategy based on compensated size to minimize the space amplification of index LSM-trees. Extensive experiments show that Scavenger significantly improves write performance and achieves lower space amplification than other KV-separated LSM-trees (including BlobDB, Titan, and TerarkDB).

关键词： Data engineering Compaction Indexes Optimization

来源：评论

学校读者我要写书评

暂无评论

Deep Q-learning Sampling Based on Advantages 5

Deep Q-learning Sampling Based on Advantages

引用

5th International Conference on Intelligent Robotics and Control engineering, IRCE 2022

作者： Xie, Ming Ren, Xinrui Yu, Jianbo Shu, Feng Shanghai Engineering Research Center of Ultra-Precision Motion Control and Measurement Academy for Engineering and Technology Fudan University Shanghai China School of Optical-Electrical and Computer Engineering University of Shanghai for Science and Technology Shanghai China School of Microelectronics Fudan University State Key Laboratory of ASIC and System Shanghai200433 China

ISBN: (数字)9781665469951

ISBN: (纸本)9781665469951

Deep Q-learning (DQN) has shown recent success on a wide range of complicated sequential decision-making issues, especially in the classic control area. However, in most DQN training, the sampling policies, particularly the ϵ-greedy policy and its derivation rules, produce noise erroneously due to the unsteadiness and unreliability of Q values utilized for searching for optimum actions. We begin with a fundamental hypothesis: complete advantages-based exploitation may be employed to help in successful exploration, and the gap between advantages on outstanding and poor actions tends to extend during the early phases of training before stabilizing. We then present a novel DQN architecture with a superior sampling policy that is totally and directly based on advantage values. This new DQN is named A-sampling DQN, and it requires minor code changes based on the original DQN codes. In most cases, A-sampling DQN and its derivatives outperformed the traditional DQN method. © 2022 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Robust Output Feedback Control Design for Inertia Emulation by Wind Turbine Generators

Robust Output Feedback Control Design for Inertia Emulation ...

引用

IEEE General Meeting Power& Energy Society

作者： Samaneh Morovati Yichen Zhang Seddik Djouadi Kevin Tomsovic Andrew Wintenberg Mohammed Olama Electrical Engineering and Computer Science The University of Tennessee Knoxville Energy system devision Argonne National Laboratory Electrical Engineering and Computer Science University of Michigan Ann Arbor Computational Sciences and Engineering Division Oak Ridge National Laboratory

ISBN: (数字)9781665408233

ISBN: (纸本)9781665408240

Wind generation has gained widespread use as a renewable energy source. Most wind turbines and other renewables connected to the grid through converters result in a reduction in the natural inertial response to grid frequency changes. The doubly-fed induction generator (DFIG) can be controlled to compensate for this reduction and, in fact, provide faster response than traditional synchronous machines. This paper proposes to design observer based output feedback linear quadratic regulator (LQR) and H∞ control laws to realize the inertia emulation function and deliver fast frequency support. The aim is to track the reference speed by a diesel synchronous generator (DSG) in order to reach the desired inertia. The control signal is computed based on a reduced order model using the balanced truncation technique. A comparison with selective modal analysis (SMA) and balanced truncation model reduction techniques is presented. Comprehensive results show the effective emulation of synthetic inertia by implementing the control laws on a nonlinear three-phase diesel-wind system. The proposed technique is analyzed for different short circuit ratio (SCR) scenarios.

关键词： Wind energy generation Renewable energy sources Regulators Modal analysis Emulation Observers Reduced order systems

来源：评论

学校读者我要写书评

暂无评论

SSC- l0 : Sparse Subspace Clustering with the l0 Inequality Constraint 7th

SSC- l0 : Sparse Subspace Clustering with the l0 Inequality...

引用

7th Asian Conference on Pattern Recognition, ACPR 2023

作者： Wang, Yangbo Zhou, Jie Lin, Qingshui Lu, Jianglin Gao, Can National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen518060 China College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China SZU Branch Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen518060 China Basic Teaching Department Liaoning Technical University Huludao125105 China College of Engineering Northeastern University BostonMA02115 United States

ISBN: (纸本)9783031476648

Self-expression learning methods often obtain a coefficient matrix to measure the similarity between pairs of samples. However, directly using all points to represent a fixed sample in a class under the self-expression framework may not be ideal, as points from other classes participate in the representing process. To alleviate this issue, this study attempts to achieve representation learning between points only coming from the same class. In practice, it is easier for data points from the same class to represent each other than that from different classes. So, when reconstructing a point, if the number of non-zero elements in the coefficient vector is limited, a model is more likely to select data points from the class where the reconstructed point lies to complete the reconstruction work. Based on this idea, we propose Sparse Subspace Clustering with the l0 inequality constraint (SSC- l0 ). In SSC- l0, the l0 inequality constraint determines the maximum number of non-zero elements in the coefficient vector, which helps SSC- l0 to conduct representation learning among the points in the same class. After introducing the simplex constraint to ensure the translation invariance of the model, an optimization method concerning l0 inequality constraint is formed to solve the proposed SSC- l0, and its convergence is theoretically analyzed. Extensive experiments on well-known datasets demonstrate the superiority of SSC- l0 compared to several state-of-the-art methods. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

The Knowledge of Cyber-Security Vulnerabilities in an Institution of Higher and University Education. A Case of ISP-Bukavu (Institut Supérieur Pédagogique de Bukavu) (TTC = Teachers’ Training College)

引用

Journal of computer and Communications 2023年第4期11卷 12-32页

作者： Dominique Wasso Kiseki Vincent Havyarimana Therence Niyonsaba Désiré Lumonge Zabagunda Walumbuka Ilundu Wail Thabo Semong Computer Engineering University of Burundi (UB) Bujumbura Burundi Ecole Normale Supérieure (ENS) Bujumbura Burundi University Research Laboratory in Modeling and Applied Statistical Engineering (LURMISTA) Nyamugerera Bujumbura Department of Physics and Technology ISP Bukavu Democratic Republic of Congo Department of Management Computer Science ISP Bukavu Democratic Republic of Congo Departement of Computer science and Information System Bostwana International University of Science and Technology Palapye Bostwana

This study pursues the objective of analyzing and verifying the knowledge of the agents of the Institut Supérieur Pédagogique/ISP-Bukavu (TTC = Teachers’ training College) in relation to the practical flaws resulting from the lack of knowledge of the observable rules in information system security. In a clearer way, it aims to verify the level of knowledge of the vulnerabilities, to verify the level of use of the antivirus software, to analyze the frequency of use of Windows update, the use of an anti-spyware software as well as a firewall software on the computer. Through a survey conducted on a sample of 100 agents of the Institut Supérieur Pédagogique/ISP-Bukavu (TTC = Teachers’ training College), the results revealed that 48% of the sample has no knowledge on computer vulnerabilities;for the use of antivirus software: 47% do not use the antivirus;for Windows update: 29% never update the Windows operating system;for anti-spyware: 48% never use;for the firewall: 50% are not informed. In fine, our results proposed a protection model VMAUSP (Vulnerability Measurability Measures Antivirus, Update, Spyware and Firewall) to users based on the behavioral approach, learning how the model works.

关键词： Threats Vulnerabilities Cybersecurity Anti-Spyware Firewall Cyber-Attack Antivirus

来源：评论

学校读者我要写书评

暂无评论

Outsourced Secure Cross-Modal Retrieval based on Secret Sharing for Lightweight Clients

引用

IEEE Internet of Things Journal 2025年

作者： Niu, Ziyu Wang, Hao Li, Zhi Su, Ye Xu, Lijuan Zhang, Yudi Susilo, Willy Shandong Normal University School of Information Science and Engineering Jinan China University of Wollongong School of Computing and Information Technology Australia Qilu University of Technology Shandong Academy of Sciences Key Laboratory of Computing Power Network and Information Security Ministry of Education National Supercomputer Center in Jinan Shandong Computer Science Center Jinan China Shandong Fundamental Research Center for Computer Science Shandong Provincial Key Laboratory of Industrial Network and Information System Security Jinan China

Cross-modal retrieval is a technique that uses one modality to query another modality in multimedia data (e.g., retrieving images based on text, or retrieving text based on images). It can break down the barriers between different modalities and achieve seamless information connection. Secure cross-modal retrieval focuses on privacy issues in cross-modal retrieval, including private data of data owners and private query requests of users. Current work on secure cross-modal retrieval protects private information through homomorphic encryption, which makes the efficiency of the retrieval phase not ideal. Therefore, the conflict between retrieval efficiency and security has become an important issue that needs to be resolved in secure cross-modal retrieval. We propose a scheme to achieve secure cross-modal retrieval in the form of secret sharing in the IoT environment. In the scheme, the data owner (DO) can secretly divide all the original data into two parts and upload them to two non-collusive cloud servers respectively. The servers store the data and provide cross-modal retrieval for users. The security of the scheme is proved under semi-honest model, and the experiments show that our scheme is more efficient than previous work in the search phase. When the query dimension is 512 and the number of latent factors is 500, the search time is reduced by more than half compared with previous work. © 2014 IEEE.

关键词： Sensitive data

来源：评论

学校读者我要写书评

暂无评论

Efficient Hybrid Multi-Population Genetic Algorithm for Multi-UAV Task Assignment in Consumer Electronics Applications

引用

IEEE Transactions on Consumer Electronics 2025年

作者： Bai, Xiaoshan Jiang, Haoyu Li, Chao Ullah, Inam Dabel, Maryam M. Al Bashir, Ali Kashif Wu, Zongze Sam, Shuzhi Shenzhen University College of Mechatronics and Control Engineering National Engineering Laboratory for Big Data System Computing Technology Shenzhen518060 China Shenzhen University College of Mechatronics and Control Engineering Shenzhen518060 China Shenzhen University Shenzhen City Joint Laboratory of Autonomous Unmanned Systems and Intelligent Manipulation Shenzhen518060 China Tashkent State University of Economics Department of Artificial Intelligence Tashkent100066 Uzbekistan University of Hafr Al Batin Department of Computer Science and Engineering College of Computer Science and Engineering Saudi Arabia Manchester Metropolitan University Manchester United Kingdom Shenzhen518060 China National University of Singapore Department of Electrical and Computer Engineering Singapore117576 Singapore

In recent years, as people’s living standards have improved and consumption concepts have been transformed, the demand for purchasing consumer electronics online has continued to grow, further stimulating the development of the logistics industry. Consequently, how to deliver consumer electronics to households at minimal cost has become a crucial factor that limits the development of the consumer technology industry. To tackle this problem, this paper studies the task assignment problem for multiple initially dispersed UAVs to deliver products to target locations while minimizing their total operation time. Each UAV can continuously provide delivery services to multiple target locations within its limited loading capacity and operation time. To solve this problem, we propose several hybrid multipopulation genetic algorithms. First, a novel crossover operator for the genetic algorithms is designed, through which a single parent chromosome can generate offspring individually. Second, two mutation mechanisms are performed to increase gene diversity. Third, multiple local search strategies are employed to enhance the populations’ fitness during each iteration of evolution. An improved 2-opt local search strategy is applied to optimize individual chromosomes when their similarity with the current best chromosome falls below a prescribed threshold. Alternatively, local search strategies are utilized for 1-opt, 2h-opt and interchange processes. Combining local search strategies, genetic operators, and the multi-population mechanism leads to several hybrid multi-population genetic algorithms. Numerical simulations and experimental tests demonstrate that the hybrid multi-population genetic algorithm, integrated with the improved 2-opt and 1-opt local search strategies, exhibits superior performance among the designed hybrid genetic algorithms, the minimum marginal cost algorithm (MMA), and the existing popular Co-evolutionary Multi-population Genetic Algorithm (CMGA). In exp

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Securely Fine-tuning Pre-trained Encoders Against Adversarial Examples

arXiv

引用

arXiv 2024年

作者： Zhou, Ziqi Li, Minghui Liu, Wei Hu, Shengshan Zhang, Yechao Wan, Wei Xue, Lulu Zhang, Leo Yu Yao, Dezhong Jin, Hai National Engineering Research Center for Big Data Technology and System China Services Computing Technology and System Lab. China Cluster and Grid Computing Lab. China Hubei Engineering Research Center on Big Data Security China Hubei Key Laboratory of Distributed System Security China School of Computer Science and Technology Huazhong University of Science and Technology China School of Software Engineering Huazhong University of Science and Technology China School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Information and Communication Technology Griffith University Australia

With the evolution of self-supervised learning, the pre-training paradigm has emerged as a predominant solution within the deep learning landscape. Model providers furnish pre-trained encoders designed to function as versatile feature extractors, enabling downstream users to harness the benefits of expansive models with minimal effort through fine-tuning. Nevertheless, recent works have exposed a vulnerability in pre-trained encoders, highlighting their susceptibility to downstream-agnostic adversarial examples (DAEs) meticulously crafted by attackers. The lingering question pertains to the feasibility of fortifying the robustness of downstream models against DAEs, particularly in scenarios where the pre-trained encoders are publicly accessible to the attackers. In this paper, we initially delve into existing defensive mechanisms against adversarial examples within the pre-training paradigm. Our findings reveal that the failure of current defenses stems from the domain shift between pre-training data and downstream tasks, as well as the sensitivity of encoder parameters. In response to these challenges, we propose Genetic Evolution-Nurtured Adversarial Fine-tuning (Gen-AF), a two-stage adversarial fine-tuning approach aimed at enhancing the robustness of downstream models. Gen-AF employs a genetic-directed dual-track adversarial fine-tuning strategy in its first stage to effectively inherit the pre-trained encoder. This involves optimizing the pre-trained encoder and classifier separately while incorporating genetic regularization to preserve the model’s topology. In the second stage, Gen-AF assesses the robust sensitivity of each layer and creates a dictionary, based on which the top-k robust redundant layers are selected with the remaining layers held fixed. Upon this foundation, we conduct evolutionary adaptability fine-tuning to further enhance the model’s generalizability. Our extensive experiments, conducted across ten self-supervised training methods and six

关键词： Topology

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：