检索结果-内蒙古大学图书馆

9th International Conference on Computer and Communications, ICCC 2023

作者： Cai, Wenjing Wang, Fei Gao, Lipeng Northwestern Polytechnical University School of Cybersecurity Xi'an China State Key Laboratory of Geo-Information Engineering Xi'an China Northwestern Polytechnical University School of Software Xi'an China

ISBN: (纸本)9798350317251

In response to the challenges faced in acquiring datasets for building defect segmentation algorithms, and the narrow, elongated nature of defect areas that can contain a variety of composite defects, rendering existing segmentation networks ineffective, an attention-mechanism-based human feedback reinforcement learning algorithm for building defect segmentation has been proposed. The design of the reward network is first carried out, and the information extraction ability of the reward network is enhanced by introducing a deep mixed multi-scale attention mechanism, thus reducing the number of iterations. Subsequently, by improving the deep deterministic policy gradient, human feedback is integrated into the reinforcement learning framework for the overall algorithm design. Experimental results demonstrate that, compared with existing supervised segmentation networks, the proposed algorithm reduces the per-pixel cross-entropy loss by 3.52%, and compared with unsupervised segmentation networks, it reduces this loss by 7.43%. In addition, the algorithm exhibits superior results in terms of recall, precision, and other metrics. © 2023 IEEE.

关键词： Defects

来源：评论

学校读者我要写书评

暂无评论

A formal model for multiagent Q-learning on graphs

引用

Science China(Information Sciences) 2025年

作者： Jinzhuo LIU Guangchen JIANG Chen CHU Yong LI Zhen WANG Shuyue HU School of Software Yunnan University School of Artificial Intelligence Optics and Electronics (iO PEN)Northwestern Polytechnical University Department of Statistics School of Statistics and MathematicsYunnan University of Finance and Economics Yunnan Key Laboratory of Service Computing Yunnan University of Finance and Economics School of Information Science&Engineering Yunnan University School of Cybersecurity Northwestern Polytechnical University Shanghai Artificial Intelligence Laboratory

Understanding the dynamics of multi-agent learning has long been an important research topic. Existing research has focused mostly on 2-agent games or well-mixed populations. However, in real-world multi-agent systems, agents often interact in spatially or socially structured networks(or graphs). In this paper, we examine the dynamics of multi-agent Q-learning on graphs. Combining mean-field theory and combinatorics analysis, we present a new analytical approach to formally describe the time evolution of Q-values in the system with a topological structure. Through extensive numerical simulations, we show that our theory consistently provides an accurate depiction of the Q-learning dynamics across different typical games, initial conditions, and various graph structures, encompassing regular graphs, scale-free graphs, and random graphs. Moreover, we show that when comparing regular graphs to other types of graphs with the same average degree, the differences in the system evolution are largely attributed to the behaviors and Q-values of agents with lower degrees.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Not All Fabrics Are Created Equal: Exploring eFPGA Parameters for IP Redaction

引用

IEEE Transactions on Very Large Scale Integration (VLSI) Systems 2023年第10期31卷 1459-1471页

作者： Bhandari, Jitendra Moosa, Abdul Khader Thalakkattu Tan, Benjamin Pilato, Christian Gore, Ganesh Tang, Xifan Temple, Scott Gaillardon, Pierre-Emmanuel Karri, Ramesh New York University Center for Cybersecurity New YorkNY11201 United States University of Calgary Department of Electrical and Software Engineering CalgaryT2N 1N4 Canada Politecnico di Milano Informazione e Bioingegneria Dipartimento di Elettronica Milan20133 Italy The University of Utah Laboratory for NanoIntegrated Systems Electrical and Computer Engineering Department Salt Lake CityUT84112 United States

Semiconductor design houses rely on third-party foundries to manufacture their integrated circuits (ICs). While this trend allows them to tackle fabrication costs, it introduces security concerns as external (and potentially malicious) parties can access critical parts of the designs and steal or modify the intellectual property (IP). Embedded field-programmable gate array (eFPGA) redaction is a promising technique to protect critical IPs of an ASIC by redacting (i.e., removing) critical parts and mapping them onto a custom reconfigurable fabric. Only trusted parties will receive the correct bitstream to restore the redacted functionality. While previous studies imply that using an eFPGA is a sufficient condition to provide security against IP threats like reverse-engineering, whether this truly holds for all eFPGA architectures is unclear, thus motivating the study in this article. We examine the security of eFPGA fabrics generated by varying different FPGA design parameters. We characterize the power, performance, and area (PPA) characteristics and evaluate each fabric's resistance to Boolean satisfiability (SAT)-based bitstream recovery. Our results encourage designers to work with custom eFPGA fabrics rather than off-The-shelf commercial FPGAs and reveals that only considering a redaction fabric's bitstream size is inadequate for gauging security. © 1993-2012 IEEE.

关键词： Fabrication

来源：评论

学校读者我要写书评

暂无评论

Reversible Data Hiding With Secret Encrypted Image Sharing and Adaptive Coding

引用

IEEE Internet of Things Journal 2025年第13期12卷 23933-23945页

作者： Fang, Guangtian Wang, Feng Zhao, Chenbin Qin, Chuan Chang, Ching-Chun Chang, Chin-Chen Fujian University of Technology College of Computer Science and Mathematics Fuzhou350118 China Fujian University of Technology College of Computer Science and Mathematics Key Laboratory of Big Data Mining and Applications Fuzhou350118 China Monash University Faculty of IT Department of Software Systems and Cybersecurity VIC3800 Australia Ministry of Education School of Cyber Science and Engineering Key Laboratory of Aerospace Information Security and Trusted Computing Wuhan430072 China University of Shanghai for Science and Technology School of Optical-Electrical and Computer Engineering Shanghai200093 China Feng Chia University Information and Communication Security Research Center Taichuang 40724 Taiwan Feng Chia University Department of Information Engineering and Computer Science Taichung40724 Taiwan

To ensure the security of image information and facilitate efficient management in the cloud, the utilization of reversible data hiding in encrypted images (RDHEI) has emerged as pivotal. However, most existing RDHEI schemes suffer from lower security and limited embedding capacity. To tackle these challenges, we propose a reversible data hiding with secret encrypted image sharing and adaptive coding scheme. Specifically, in the encryption phase, we introduce an improved secret sharing (SS) encryption method based on the chinese remainder theorem for polynomials (CRTP). This method not only improves the security of encrypted images but also vacates a larger room for embedding. In the embedding phase, we introduce an adaptive coding embedding approach using XOR preservation (XORP) and huffman coding, which provides high embedding capacity. Experimental results and security analysis demonstrate that our proposed encryption method achieves optimal values in security indicators for encrypted images, such as information entropy, histograms, NPCR and UACI. The proposed embedding method is superior to some state-of-the-art schemes in terms of embedding capacity. Furthermore, in datasets BOSSBase and BOWS2, the average embedding rates of the proposed embedding approach can reach 2.1745 bpp and 2.0656 bpp, respectively. © 2025 IEEE.

关键词： Cryptography Encryption Security Data mining Adaptive coding Correlation Cloud computing Polynomials Image coding Streaming media

来源：评论

学校读者我要写书评

暂无评论

Verification Tool for Securing RISC-V FPGA-Based Process Control System

Verification Tool for Securing RISC-V FPGA-Based Process Con...

引用

Intermountain engineering, Technology and Computing (IETC)

作者： Christine Johnson Ann Marie Reinhold Clemente Izurieta Bradley M. Whitaker Brock LaMeres Software Engineering and Cybersecurity Laboratory Montana State University Bozeman MT USA Pacific Northwest National Laboratory Richland WA USA Idaho National Laboratory Idaho Falls ID USA Electrical & Computer Engineering Department Montana State University Bozeman MT USA

ISBN: (数字)9798331512828

ISBN: (纸本)9798331512835

Industrial Control Systems (ICSs) are a crucial component of critical infrastructure and a popular target for cyberattacks. While most research focuses on the defense of large-scale networks of ICSs, it is critical to expand research on the small-scale networks of Process Control Systems (PCSs), which attackers may target to remain undetectable from the security of the larger network. One potential method of protection for PCS networks is the use of verification tools; however, existing research on verification tools focuses on the detection of attacks without mitigation. This research describes two instantiations of a verification tool for a PCS controlled by a RISC-V computer implemented on an FPGA. The first version of the verification tool provides passive detection, informing an engineering station of an attack, and the second version provides detection with mitigation against unauthorized command messages and malicious software downloads. An experimental testbed was developed to test both versions, consisting of: (1) a circuit to charge a battery; (2) an FPGA controller that controls charge/discharge based on user input; (3) an engineering station that provides control data and updates firmware to the FPGA; and (4) a verification tool that verifies input forwarded by a passive serial tap, connected through the FPGA’s hardware. The experimental data yielded promising results, with the tool successfully providing mitigation and detection against attacks on the serial communication channel between the engineering station and the FPGA. This approach has applicability to common ICS computer devices, such as programmable logic controllers (PLCs).

关键词： Reduced instruction set computing Prevention and mitigation Programmable logic devices Process control Logic gates Malware Protection Field programmable gate arrays Monitoring Microprogramming

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Mapping of Quantitative Measures to Qualitative Characteristics in Hierarchical software Quality Assurance

Unsupervised Mapping of Quantitative Measures to Qualitative...

引用

Intermountain engineering, Technology and Computing (IETC)

作者： Kaveen Liyanage Ethan L. Gerard Derek Reimanis Ann Marie Reinhold Clemente Izurieta Brock J. LaMeres Bradley M. Whitaker Department of Electrical and Computer Engineering Montana State University Bozeman MT USA Software Engineering and Cybersecurity Laboratory Montana State University Bozeman MT USA Pacific Northwest National Laboratory Richland WA USA Idaho National Laboratory Idaho Falls ID USA

ISBN: (数字)9798331512828

ISBN: (纸本)9798331512835

Ensuring software quality is essential to the software development and deployment process. Evaluating software for known vulnerabilities and weaknesses is a method of assessing the quality of software. However, the existence of a large number of vulnerabilities and weaknesses can hinder decision-making and mitigation. To address this, hierarchical software quality models aggregate quantitative weakness measures into qualitative characteristics to simplify decision making. However, challenges exist when mapping quantitative measures to qualitative characteristics, especially when the relationship between measures and characteristics is ill-defined/unknown or when prior knowledge is absent, thus posing threats to the construct validity of the mapping. This paper presents a pseudo-label-based regression framework to generate qualitative values when given a set of quantitative measurements. To exemplify this research, we created a use case where we applied the framework to binary program analysis.

关键词： Quality assurance Prevention and mitigation Decision making software quality Manuals Machine learning Predictive models Benchmark testing software measurement software development management

来源：评论

学校读者我要写书评

暂无评论

Enhancing Architectural Information Extraction Using DHAM-based Segmentation Network

Enhancing Architectural Information Extraction Using DHAM-ba...

引用

International Conference on Computer and Communications (ICCC)

作者： Wenjing Cai Fei Wang Lipeng Gao School of Cybersecurity Northwestern Polytechnical University Xi'an China State Key Laboratory of Geo-Information Engineering Xi'an China School of Software Northwestern Polytechnical University Xi'an China

关键词：

来源：评论

学校读者我要写书评

暂无评论

GraphMU: Repairing Robustness of Graph Neural Networks via Machine Unlearning

arXiv

引用

arXiv 2024年

作者： Wu, Tao Cao, Xinwen Wang, Chao Qiao, Shaojie Xian, Xingping Yuan, Lin Cui, Canyixing Liu, Yanbing The School of Cybersecurity and Information Law Chongqing University of Posts and Telecommunications Chongqing400065 China The School of Software Engineering Chengdu University of Information Technology Chengdu610225 China The School of Computer and Information Science Chongqing Normal University Chongqing401331 China Yunnan Key Laboratory of Software Engineering Yunnan650504 China

Graph Neural Networks (GNNs) have demonstrated significant application potential in various fields. However, GNNs are still vulnerable to adversarial attacks. Numerous adversarial defense methods on GNNs are proposed to address the problem of adversarial attacks. However, these methods can only serve as a defense before poisoning, but cannot repair poisoned GNN. Therefore, there is an urgent need for a method to repair poisoned GNN. In this paper, we address this gap by introducing the novel concept of model repair for GNNs. We propose a repair framework, Repairing Robustness of Graph Neural Networks via Machine Unlearning (GraphMU), which aims to fine-tune poisoned GNN to forget adversarial samples without the need for complete retraining. We also introduce a unlearning validation method to ensure that our approach effectively forget specified poisoned data. To evaluate the effectiveness of GraphMU, we explore three fine-tuned subgraph construction scenarios based on the available perturbation information: (i) Known Perturbation Ratios, (ii) Known Complete Knowledge of Perturbations, and (iii) Unknown any Knowledge of Perturbations. Our extensive experiments, conducted across four citation datasets and four adversarial attack scenarios, demonstrate that GraphMU can effectively restore the performance of poisoned GNN. Copyright © 2024, The Authors. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Non-profiled Deep-Learning-Based Power Analysis of the SM4 and DES Algorithms

引用

Chinese Journal of Electronics 2021年第3期30卷 500-507页

作者： MA Xiangliang LI Bing WANG Hong WU Di ZHANG Lizhen HUANG Kezhen DUAN Xiaoyi School of Integrated Circuits Tsinghua University National Research Center for Information Technology Security China Cybersecurity Review Technology and Certification Center Trusted Computing and Information Assurance Laboratory Institute of Software Chinese Academy of Sciences University of Chinese Academy of Sciences Department of Electronics and Information Engineering Beijing Electronic Science and Technology Institute

Power analysis methods are commonly used for evaluating the security of cryptographic *** are characteristically low-cost and display a high success rate and the ability to obtain important device information, e.g., keys. Given the current wide application of deep-learning technology, there is a growing tendency to incorporate power-analysis technology in *** study investigates non-profiled deep-learning-based power analysis. The labels used in this attack are uncertain,and the attack conditions required are greatly reduced. We choose the Recurrent neural network(RNN), multilayer perceptron, and convolutional neural network algorithms,which use the same network structure, to recover the keys for the SM4 software and DES hardware *** propose combining the RNN algorithm with power analysis, and validate the benefits experimentally. The experimental results show that they all successfully recover the correct key for the SM4 software implementation,although the RNN algorithm by itself achieves a better effect. This conclusion also applies to attacks on the DES hardware implementation but is limited to labels based on the bit model.

关键词： cryptography nonprofiled deep-learning-based power analysis cryptographic device security recurrent neural nets RNN algorithm deep learning (artificial intelligence) DES hardware implementation convolutional neural nets convolutional neural network algorithms SM4 software implementation multilayer perceptrons recurrent neural network multilayer perceptron

来源：评论

学校读者我要写书评

暂无评论

The Role of Far-Sighted Agents on the Evolution of Cooperation in Social Dilemma

SSRN

引用

SSRN 2023年

作者： Shi, Juan Liu, Xucheng Li, Jiqin Shu, Youqi Wang, Zhen Liu, Jinzhuo School of Automation Northwestern Polytechnical University Shaanxi Xi’an710072 China School of Software Yunnan University Yunnan Kunming650504 China The Key Laboratory for Software Engineering of Yunnan Province Yunnan Kunming650504 China School of Cybersecurity Northwestern Polytechnical University Shaanxi Xi’an710072 China

Why would natural selection facilitate the prevailing cooperation in selfish groups of individuals? A common framework for studying this problem is evolutionary game theory. Here, we propose a new strategy update rule, called the far-Sighted update rule, that enables agents to take into account the possible responses of their neighbors, which means a far-Sighted agent does not have a short-sighted, static view of the environment but rather a long-term view that considers environmental changes. We study the effect of the far-Sighted update rule on behavioral evolution in two cases. The level of cooperation can be significantly promoted for populations of myopic agents acting with a certain probability with the far-Sighted update rule. The cooperation rate of the population did not drop below a threshold even under very hostile conditions. Moreover, the stronger the intelligence's willingness to consider long-term views is, the larger this threshold will be. The theoretical results of the mean field are consistent with the simulation results. However, for far-Sighted agents, which act exactly as far-Sighted update rule, it shows sensitivity to the network structure. In a mixed population of far-Sighted agents and myopic agents, even if all myopic agents become defectors, the far-Sighted agents can still maintain a certain cooperator ratio. Thus, our work can help to understand the evolution of populations of agents with a higher intelligence that can consider environmental changes. © 2023, The Authors. All rights reserved.

关键词： Mean field theory

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：