检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Zhou, Ziqi Hu, Shengshan Zhao, Ruizhi Wang, Qian Zhang, Leo Yu Hou, Junhui Jin, Hai School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Computer Science and Technology Huazhong University of Science and Technology China School of Cyber Science and Engineering Wuhan University China School of Information and Communication Technology Griffith University Australia Department of Computer Science City University of Hong Kong Hong Kong National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Hubei Key Laboratory of Distributed System Security China Hubei Engineering Research Center on Big Data Security China Cluster and Grid Computing Lab

Self-supervised learning usually uses a large amount of unlabeled data to pre-train an encoder which can be used as a general-purpose feature extractor, such that downstream users only need to perform fine-tuning operations to enjoy the benefit of "large model". Despite this promising prospect, the security of pre-trained encoder has not been thoroughly investigated yet, especially when the pre-trained encoder is publicly available for commercial use. In this paper, we propose AdvEncoder, the first framework for generating downstream-agnostic universal adversarial examples based on the pre-trained encoder. AdvEncoder aims to construct a universal adversarial perturbation or patch for a set of natural images that can fool all the downstream tasks inheriting the victim pre-trained encoder. Unlike traditional adversarial example works, the pre-trained encoder only outputs feature vectors rather than classification labels. Therefore, we first exploit the high frequency component information of the image to guide the generation of adversarial examples. Then we design a generative attack framework to construct adversarial perturbations/patches by learning the distribution of the attack surrogate dataset to improve their attack success rates and transferability. Our results show that an attacker can successfully attack downstream tasks without knowing either the pre-training dataset or the downstream dataset. We also tailor four defenses for pre-trained encoders, the results of which further prove the attack ability of AdvEncoder. Our codes are available at: https://github. com/CGCL-codes/AdvEncoder. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

DarkFed: A Data-Free Backdoor Attack in Federated Learning

arXiv

引用

arXiv 2024年

作者： Li, Minghui Wan, Wei Ning, Yuxuan Hu, Shengshan Xue, Lulu Zhang, Leo Yu Wang, Yichen School of Software 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 School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Computer Science and Technology Huazhong University of Science and Technology China School of Information and Communication Technology Griffith University Australia

Federated learning (FL) has been demonstrated to be susceptible to backdoor attacks. However, existing academic studies on FL backdoor attacks rely on a high proportion of real clients with main task-related data, which is impractical. In the context of real-world industrial scenarios, even the simplest defense suffices to defend against the state-of-the-art attack, 3DFed. A practical FL backdoor attack remains in a nascent stage of development. To bridge this gap, we present DarkFed. Initially, we emulate a series of fake clients, thereby achieving the attacker proportion typical of academic research scenarios. Given that these emulated fake clients lack genuine training data, we further propose a data-free approach to backdoor FL. Specifically, we delve into the feasibility of injecting a backdoor using a shadow dataset. Our exploration reveals that impressive attack performance can be achieved, even when there is a substantial gap between the shadow dataset and the main task dataset. This holds true even when employing synthetic data devoid of any semantic information as the shadow dataset. Subsequently, we strategically construct a series of covert backdoor updates in an optimized manner, mimicking the properties of benign updates, to evade detection by defenses. A substantial body of empirical evidence validates the tangible effectiveness of DarkFed. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

An Adaptive Early Warning Method of Cascading Failures Caused by Coordinated Cyber-attacks

引用

CSEE Journal of Power and Energy Systems 2023年

作者： Wang, Yufei Li, June Qiu, Jian Chen, Yangrong Key Laboratory of Aerospace Information Security and Trusted Computing Ministry of Education School of Cyber Science and Engineering Wuhan University Hubei Wuhan430072 China Global Energy Interconnection Research Institute Co. Ltd Beijing102209 China Electric Power Planning & Engineering Institute Xicheng District Beijing100120 China

In order to accurately early warning the cascading failures caused by coordinated cyber-attacks (CFCC) in grid cyber-physical systems (GCPS), an adaptive early warning method of CFCC is proposed. Firstly, the evolutionary mechanism of CFCC is analyzed from the attackers" view, the CFCC mathematical model is established, and the transition processes of GCPS running states under the influence of CFCC staged failures are discussed. Then, the mathematical model of the adaptive early warning method is established. Further, the mathematical model of the adaptive early warning method is mapped as an adaptive control process with tolerating staged failures damage, and the solving process is presented to infer the CFCC and its next evolution trend. And the decision-making idea of the optimal active defense scheme is proposed considering the costs and gains of various defense measures. Finally, to verify the effectiveness of the adaptive early warning method, the warning and defense processes of a typical CFCC are simulated in GCPS experimental system based on CEPRI-36 BUS. © 2023 China Electric Power Research Institute. All rights reserved.

关键词： Network security

来源：评论

学校读者我要写书评

暂无评论

ON SHRINKING TARGETS FOR LINEAR EXPANDING AND HYPERBOLIC TORAL ENDOMORPHISMS

arXiv

引用

arXiv 2024年

作者： Hu, Zhang-Nan Persson, Tomas Wu, Wanlou Zhang, Yiwei College of Science China University of Petroleum Beijing102249 China Centre for Mathematical Sciences Lund University Box 118 Lund221 00 Sweden School of Mathematics and Statistics Jiangsu Normal University Xuzhou221116 China School of Mathematics and Statistics Center for Mathematical Sciences Hubei Key Laboratory of Engineering Modeling and Scientific Computing Huazhong University of Science and Technology Wuhan430074 China

Let A be an invertible d × d matrix with integer elements. Then A determines a self-map T of the d-dimensional torus Td = Rd/Zd. Given a real number τ > 0, and a sequence {zn} of points in Td, let Wτ be the set of points x ∈ Td such that Tn(x) ∈ B(zn, e−nτ) for infinitely many n ∈ . The Hausdorff dimension of Wτ has previously been studied by Hill-Velani and Li-Liao-Velani-Zorin. We provide complete results on the Hausdorff dimension of Wτ for any expanding matrix. For hyperbolic matrices, we compute the dimension of Wτ only when A is a 2 × 2 matrix. We give counterexamples to a natural candidate for a dimension formula for general dimension *** Codes 37C45, 37E05 Copyright © 2024, The Authors. All rights reserved.

关键词： Matrix algebra

来源：评论

学校读者我要写书评

暂无评论

Exploring Multi-loss Function Based Attention Framework for Vehicle Re-Identification

Exploring Multi-loss Function Based Attention Framework for ...

引用

IEEE International Symposium on Information (IT) in Medicine and Education, ITME

作者： Wei Liu Yi Zhang Zhiming Luo Shaozi Li East China Jiaotong University Jiangxi China Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering Yi Chang China Postdoc Center of Information and Communication Engineering Xiamen University China The Key Laboratory of Cognitive Computing and Intelligent Information Processing of Fujian Education Institutions Wuyishan China Cognitive Science Department Xiamen University Fujian China

Vehicle Re-identification(Re-ID) refers to the task of matching a specific vehicle across multiple non-overlapping cameras. Precise vehicle Re-ID suffers from two main challenges: (1) different vehicle identities with a similar appearance of the same type have a subtle inter-instance discrepancy, (2) one vehicle usually has large intra-instance differences due to viewpoint and illumination variations. To address these challenges, in this paper, we proposed an Attention-based Framework for Vehicle Re-ID named AFVR, which combines the benefits of both the cross-entropy loss function and the triplet loss function. Experimental results on two popular vehicle Re-ID benchmarks demonstrate the effectiveness of our proposed method, compared with the state of the arts.

关键词： Feature extraction Cameras Training Measurement Lighting Visualization Benchmark testing

来源：评论

学校读者我要写书评

暂无评论

Deepstsnet: Reconstructing the Quantum State-Resolved Thermochemical Nonequilibrium Flowfield Using Deep Neural Operator Learning With Scarce Data

SSRN

引用

SSRN 2023年

作者： Lv, Jiaqi Hong, Qizhen Wang, Xiaoyong Mao, Zhiping Sun, Quanhua State Key Laboratory of High Temperature Gas Dynamics Institute of Mechanics Chinese Academy of Sciences Beijing100190 China School of Engineering Science University of Chinese Academy of Sciences Beijing100049 China School of Mathematical Sciences Fujian Provincial Key Laboratory of Mathematical Modeling and High-Performance Scientific Computing Xiamen University Xiamen361005 China

The hypersonic flow is in a thermochemical nonequilibrium state due to the high-temperature caused by the strong shock compression. In a thermochemical nonequilibrium flow, the distribution of molecular internal energy levels strongly deviates from the equilibrium distribution (i.e., the Boltzmann distribution). It is intractable to directly obtain the microscopic nonequilibrium distribution from existed experimental measurements usually described by macroscopic field variables such as temperature or velocity. Motivated by the idea of deep multi-scale multi-physics neural network (DeepMMNet) proposed in [Mao et al, J. Comput. Phys., 2021], we develop in this paper a data assimilation framework called DeepStSNet to accurately reconstruct the quantum state-resolved thermochemical nonequilibrium flowfield by using sparse experimental measurements of vibrational temperature and pre-trained deep neural operator networks (DeepONets). In particular, we first construct several DeepONets to express the coupled dynamics between field variables in the thermochemical nonequilibrium flow and to approximate the state-to-state (StS) approach, which traces the variation of each vibrational level of molecule accurately. These proposed DeepONets are then trained by using the numerical simulation data, and would later be served as building blocks for the DeepStSNet. We demonstrate the effectiveness and accuracy of DeepONets with different test cases showing that the density and energy of vibrational groups as well as the temperature and velocity fields are predicted with high accuracy. We then extend the architectures of DeepMMNet by considering a simplified thermochemical nonequilibrium model, i.e., the 2T model, showing that the entire thermochemical nonequilibrium flowfield is well predicted by using scattered measurements of full or even partial field variables. We next consider a more accurate and complex thermochemical nonequilibrium model, i.e., the StS-CGM model, and develop a

关键词： Multiphysics

来源：评论

学校读者我要写书评

暂无评论

Robin: A Novel Method to Produce Robust Interpreters for Deep Learning-Based Code Classifiers

arXiv

引用

arXiv 2023年

作者： Li, Zhen Zhang, Ruqian Zou, Deqing Wang, Ning Li, Yating Xu, Shouhuai Chen, Chen Jin, Hai National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Hubei Key Laboratory of Distributed System Security Hubei Engineering Research Center on Big Data Security Cluster and Grid Computing Lab China School of Cyber Science and Engineering Huazhong University of Science and Technology Wuhan430074 China Department of Computer Science University of Colorado Colorado Springs United States Center for Research in Computer Vision University of Central Florida United States School of Computer Science and Technology Huazhong University of Science and Technology Wuhan430074 China

Deep learning has been widely used in source code classification tasks, such as code classification according to their functionalities, code authorship attribution, and vulnerability detection. Unfortunately, the black-box nature of deep learning makes it hard to interpret and understand why a classifier (i.e., classification model) makes a particular prediction on a given example. This lack of interpretability (or explainability) might have hindered their adoption by practitioners because it is not clear when they should or should not trust a classifier's prediction. The lack of interpretability has motivated a number of studies in recent years. However, existing methods are neither robust nor able to cope with out-of-distribution examples. In this paper, we propose a novel method to produce Robust interpreters for a given deep learning-based code classifier;the method is dubbed Robin. The key idea behind Robin is a novel hybrid structure combining an interpreter and two approximators, while leveraging the ideas of adversarial training and data augmentation. Experimental results show that on average the interpreter produced by Robin achieves a 6.11% higher fidelity (evaluated on the classifier), 67.22% higher fidelity (evaluated on the approximator), and 15.87x higher robustness than that of the three existing interpreters we evaluated. Moreover, the interpreter is 47.31% less affected by out-of-distribution examples than that of LEMNA. Copyright © 2023, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

An adjoint-free algorithm for conditional nonlinear optimal perturbations (CNOPs) via sampling

arXiv

引用

arXiv 2022年

作者： Shi, Bin Sun, Guodong State Key Laboratory of Scientific and Engineering Computing Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing100190 China State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics Institute of Atmospheric Physics Chinese Academy of Sciences Beijing100029 China University of Chinese Academy of Sciences Beijing100049 China

In this paper, we propose a sampling algorithm based on state-of-the-art statistical machine learning techniques to obtain conditional nonlinear optimal perturbations (CNOPs), which is different from traditional (deterministic) optimization methods.1 Specifically, the traditional approach is unavailable in practice, which requires numerically computing the gradient (first-order information) such that the computation cost is expensive since it needs a large number of times to run numerical models. However, the sampling approach directly reduces the gradient to the objective function value (zeroth-order information), which also avoids using the adjoint technique that is unusable for many atmosphere and ocean models and requires large amounts of storage. We show an intuitive analysis for the sampling algorithm from the law of large numbers and further present a Chernoff-type concentration inequality to rigorously characterize the degree to which the sample average probabilistically approximates the exact gradient. The experiments are implemented to obtain the CNOPs for two numerical models, the Burgers equation with small viscosity and the Lorenz-96 model. We demonstrate the CNOPs obtained with their spatial patterns, objective values, computation times and nonlinear error growth. Compared with the performance of the three approaches, all the characters for quantifying the CNOPs are nearly consistent, while the computation time using the sampling approach with fewer samples is extremely shorter. In other words, the new sampling algorithm shortens the computation time to the utmost at the cost of losing little accuracy. © 2022, CC BY-NC-SA.

关键词： Numerical models

来源：评论

学校读者我要写书评

暂无评论

Energy efficiency optimization of generalized spatial modulation with sub-connected hybrid precoding

arXiv

引用

arXiv 2019年

作者： Chen, Kai Yang, Jing Ge, Xiaohu Li, Yonghui Tian, Lin Shi, Jinglin School of Electronic Information and Communications Huazhong University of Science and Technology WuhanHubei China School of Electrical and Information Engineering University of Sydney Sydney Australia Beijing Key Laboratory of Mobile Computing and Pervasive Devices Institute of Computing Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences China

Energy efficiency (EE) optimization of millimeter wave (mm-Wave) massive multiple-input multiple-output (MIMO) systems is emerging as an important challenge for the fifth generation (5G) mobile communication systems. However, the power of radio frequency (RF) chains increases sharply due to the high carrier frequency in mm-Wave massive MIMO systems. To overcome this issue, a new energy efficiency optimization solution is proposed based on the structure of the generalized spatial modulation (GSM) and sub-connected hybrid precoding (HP). Moreover, the computation power of mm-Wave massive MIMO systems is considered for optimizing the EE. Simulation results indicate that the EE of the GSM-HP scheme outperforms the full digital precoding (FDP) scheme in the mm-Wave massive MIMO scene, and 88% computation power can be saved by the proposed GSM-HP scheme. Copyright © 2019, The Authors. All rights reserved.

关键词： Millimeter waves

来源：评论

学校读者我要写书评

暂无评论

Energy efficiency of generalized spatial modulation aided massive MIMO systems

arXiv

引用

arXiv 2019年

作者： Zheng, Shuang Yang, Jing Ge, Xiaohu Li, Yonghui Tian, Lin Shi, Jinglin School of Electronic Information and Communications Huazhong University of Science and Technology Wuhan Hubei China School of Electrical and Information Engineering University of Sydney Sydney Australia Beijing Key Laboratory of Mobile Computing and Pervasive Devices Institute of Computing Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences China

One of focuses in green communication studies is the energy efficiency (EE) of massive multiple-input multiple-output (MIMO) systems. Although the massive MIMO technology can improve the spectral efficiency (SE) of cellular networks by configuring a large number of antennas at base stations (BSs), the energy consumption of radio frequency (RF) chains increases dramatically. The increment of energy consumption is caused by the increase of RF chain number to match the antenna number in massive MIMO communication systems. To overcome this problem, a generalized spatial modulation (GSM) solution is presented to simultaneously reduce the number of RF chains and maintain the SE of massive MIMO communication systems. A EE model is proposed to estimate the transmission and computation power of massive MIMO communication systems with GSM. Simulation results demonstrate that the EE of massive MIMO communication systems with GSM outperforms the massive MIMO communication systems without GSM. Besides, the computation power consumed by massive MIMO communication systems with GSM is effectively reduced. Copyright © 2019, The Authors. All rights reserved.

关键词： Energy efficiency

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：