检索结果-内蒙古大学图书馆

Asian control Conference

作者： Hong Lin Nuo Chen Yang Chen Xiang Li Cong Li Department of Electronic Engineering Adaptive Networks and Control Lab School of Information Science and Technology Fudan University China Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University China The Institute of Complex Networks and Intelligent Systems Shanghai Research Institute for Intelligent Autonomous Systems Tongji University China

ISBN: (数字)9789887581598

ISBN: (纸本)9798331540845

Within online social networks, the rapid expansion of automated accounts, known as social bots, poses a significant challenge. This dynamic has sparked a strategic contest between bots and detectors. Both sides continually learn and adjust their strategies to gain an edge in this silent war. To effectively manage risks from bot evolution, proactive detection methods have garnered widespread attention. In this work, we propose a graph-based and structure-aware adversarial learning framework for multi-category social bot detection. In detail, we adopt a categorical generative adversarial network to create synthetic evolved bot samples for adversarial learning, aiming to boost the detection performance. Besides, we design a user representation learning module, including local and remote feature extractors to mine neighborhood information and achieve multiple network structure representations. Results demonstrate that our model outperforms state-of-the-art approaches and exhibits strong robustness.

关键词： Representation learning Training Technological innovation Social networking (online) Image edge detection Detectors Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Predefined-Time Bipartite Consensus of Networked Euler-Lagrange Systems via Sliding-Mode control

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS 2022年第12期69卷 4989-4993页

作者： Tao, Meng Liu, Xiaoyang Shao, Shao Cao, Jinde Jiangsu Normal Univ Sch Comp Sci & Technol Res Ctr Complex Networks & Swarm Intelligence Xuzhou 221116 Jiangsu Peoples R China Northeastern Univ Qinhuangdao Sch Control Engn Qinhuangdao 066004 Hebei Peoples R China Southeast Univ Sch Math Nanjing 210096 Peoples R China Yonsei Univ Yonsei Frontier Lab Seoul 03722 South Korea

This brief investigates the predefined-time bipartite consensus of multiple Euler-Lagrange systems with a dynamic leader. A new distributed predefined-time observer is added to estimate the desired velocity of each follower. Then, a new sliding surface and a distributed protocol are constructed to guarantee the networked systems achieve the bipartite consensus within a predefined time. It should be pointed out that, all the observing time, sliding time and reaching time can be preset in advance according to the task requirements. One numerical example with two-link robot arms is proposed to verify the effectiveness of the new designs.

关键词： Bipartite consensus Euler-Lagrange systems predefined-time consensus sliding-mode control

来源：评论

学校读者我要写书评

暂无评论

EMHDRL: Adversarial Attacks on Graph Neural networks via Hierarchical Reinforcement Learning

EMHDRL: Adversarial Attacks on Graph Neural Networks via Hie...

引用

Chinese control Conference (CCC)

作者： Guowei Sun Jie Ding Xiang Li Department of Electronic Engineering Adaptive Networks and Control Lab School of Information Science and Engineering Fudan University Shanghai China Institute of Complex Networks and Intelligent Systems Shanghai Research Institute for Intelligent Autonomous Systems Tongji University Shanghai China

Graph neural networks (GNNs), with its powerful ability to explore the topological structure of graph data, have achieved great success in various downstream tasks. However, recent studies have shown that due to the lack of robustness, GNNs are easily fooled by perturbations. For node classification tasks, the current attack methods mainly focus on manipulating edges, modifying features and adding fake nodes. Although these attacks are effective, the changes to the original graph data are easy to be detected. Therefore, to misclassify the target nodes, we propose an Edge Manipulation method via Hierarchical Deep Reinforcement Learning (EMHDRL) by taking accounting to the imperceptibility of perturbation on graph data. Through modeling the topology perturbation as Finite Horizon Markov Decision Process, EMHDRL adopts a hierarchical Q-function network to manipulate edges of graph, while trying to ensure that the degree of the attacked node and the total number of edges of the whole graph remain unchanged. Our experimental results show that the proposed method can effectively reduce the performance of GNNs in classifying the target nodes and has strong transferability.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Coevolution of opinion dynamics on evolving signed appraisal networks

引用

AUTOMATICA 2022年 137卷 110138-110138页

作者： Kang, Rongrong Li, Xiang Tongji Univ Inst Complex Networks & Intelligent Syst Shanghai Res Inst Intelligent Autonomous Syst Shanghai 201210 Peoples R China Fudan Univ Elect Engn Dept Adapt Networks & Control Lab Shanghai 200433 Peoples R China

In social networks, the expression of various opinions has great influence on appraisal networks that describe the interpersonal influence structures among agents. In this paper, we analyze opinion dynamics on signed appraisal networks, and, in turn, adjust appraisals of agents under the influence of structural balance theory and social psychological distancing. To clarify such a coevolution process, we provide mathematical formulations of the coevolution dynamics, characterize the convergence properties of opinions, and identify potential invariant appraisal structures with all initial states. We predict that the opinions of agents in a sink strongly connected component achieve a state of bipartite consensus exponentially fast, and the corresponding appraisal subnetwork converges to a fully connected structural balance state, which determines the scope of non-sinks' opinions. Finally, our findings witness their sociological significance in the context of the well-known Krackhardt's advice network and an empirical voting network. (C) 2022 Elsevier Ltd. All rights reserved.

关键词： Opinion dynamics Evolving networks Coevolution model Signed graph Structural balance theory Social distancing

来源：评论

学校读者我要写书评

暂无评论

Dynamic multi-dose vaccination with initial immunity on higher-order networks

引用

PROCEEDINGS OF THE ROYAL SOCIETY A-MATHEMATICAL PHYSICAL AND ENGINEERING SCIENCES 2024年第2300期480卷

作者： Qian, Yinuo Zhao, Dawei Xia, Chengyi Ozer, Mahmut Perc, Matjaz Tiangong Univ Sch Control Sci & Engn Tianjin 300387 Peoples R China Qilu Univ Technol Key Lab Comp Power Network & Informat Secur Minist Educ Natl Supercomp Ctr JinanShandong Acad SciShandon Jinan 250014 Peoples R China Shandong Fundamental Res Ctr Comp Sci Shandong Prov Key Lab Comp Networks Jinan 250014 Peoples R China Tiangong Univ Sch Artificial Intelligence Tianjin 300387 Peoples R China Tiangong Univ Tianjin Key Lab Intelligent Control Elect Equipmen Tianjin 300387 Peoples R China Natl Educ Culture Youth & Sports Commiss Turkish Grand Natl Assembly Ankara Turkiye Univ Maribor Fac Nat Sci & Math Maribor 2000 Slovenia Kyung Hee Univ Dept Phys Seoul 02447 South Korea Community Healthcare Ctr Dr Adolf Drolc Maribor Maribor 2000 Slovenia Complex Sci Hub Vienna A-1080 Vienna Austria

Vaccination remains crucial during pandemics for combating disease spread. We here propose a two-layer model for multi-dose vaccination policy, considering vaccine hesitancy, initial immunity, as well as media-driven information dissemination concerning vaccine strategy selection. Moreover, we consider collective opinions through social networks with higher-order interactions. We perform numerical simulations to analyse variations in vaccination strategies. We further compare obtained outcomes with the exploration of real data on vaccination behaviours. Our research indicates that information dissemination can regulate vaccination, and that models accounting for individual immunity align more closely with real-world scenarios. Furthermore, we find that considering vaccination decisions from neighbouring agents is crucial under multi-dose policies for the facilitation of faster vaccination diffusion.

关键词： multi-dose vaccination initial immunity higher-order network vaccine hesitancy

来源：评论

学校读者我要写书评

暂无评论

A Graph Transformer-Driven Approach for Network Robustness Learning

arXiv

引用

arXiv 2023年

作者： Zhang, Yu Li, Jia Ding, Jie Li, Xiang The Adaptive Networks and Control Lab Department of Electronic Engineering School of Information Science and Engineering Fudan University Shanghai200433 China The Institute of Complex Networks and Intelligent Systems Shanghai Research Institute for Intelligent Autonomous Systems Tongji University Shanghai201210 China

Learning and analysis of network robustness, including controllability robustness and connectivity robustness, is critical for various networked systems against attacks. Traditionally, network robustness is determined by attack simulations, which is very time-consuming and even incapable for large-scale networks. Network Robustness Learning, which is dedicated to learning network robustness with high precision and high speed, provides a powerful tool to analyze network robustness by replacing simulations. In this paper, a novel versatile and unified robustness learning approach based on the customized graph transformer (NRL-GT) is proposed, which consists of a particularly designed backbone and three branches to accomplish the tasks of robustness curve learning, overall robustness learning and synthetic network classification simultaneously. Numerous experiments show that: 1) NRL-GT is a unified learning framework for controllability robustness and connectivity robustness, demonstrating a strong generalization ability to ensure high precision when training and test sets are distributed differently;2) It is theoretically and experimentally demonstrated the proposed graph transformer layer is capable of outperforming classical graph neural networks. Compared to the cutting-edge methods in network robustness learning, NRL-GT can simultaneously perform network robustness learning from multiple aspects and obtains superior results in less time on both synthetic and real-world networks, especially circuit networks and power networks;3) It is worth mentioning that with the transferability of the proposed backbone, NRL-GT is able to deal with complex networks of different sizes for different tasks efficiently and effectively. Copyright © 2023, The Authors. All rights reserved.

关键词： controllability

来源：评论

学校读者我要写书评

暂无评论

Adaptive Consensus and Circuital Implementation of a Class of Faulty Multiagent Systems

引用

IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS 2022年第1期52卷 226-237页

作者： Jin, Xiao-Zheng Che, Wei-Wei Wu, Zheng-Guang Zhao, Zhen Qilu Univ Technol Sch Comp Sci & Technol Shandong Acad Sci Jinan 250353 Peoples R China Shandong Comp Sci Ctr Natl Supercomp Ctr Jinan Jinan 250014 Peoples R China Shandong Prov Key Lab Comp Networks Jinan 250014 Peoples R China Qingdao Univ Inst Complex Sci Qingdao 266071 Peoples R China Zhejiang Univ Inst Cyber Syst & Control State Key Lab Ind Control Technol Hangzhou 310027 Peoples R China Hefei Univ Technol Sch Elect Engn & Automat Hefei 230009 Peoples R China

This article is concerned with the robust adaptive fault-tolerant consensus control and the circuital implementation problems for a class of homogeneous multiagent systems with external disturbances and actuator faults. A robust adaptive consensus control strategy is developed to automatically eliminate the effects of actuator bias and partial loss-of-control-effectiveness faults, and simultaneously specify the L-2 performance of systems. The achievement of exponential consensus of the closed-loop disturbed and faulty multiagent system is provided on the basis of the Lyapunov stability theory. Furthermore, a physical implementation method is developed based on circuit theory to translate the proposed adaptive consensus control strategy into analog circuits. By using a professional tool for circuit simulations, effectiveness of the developed circuits is verified via a multiagent system composed by mobile robots with two independent driving wheels.

关键词： Actuator faults circuital implementation multiagent systems robust adaptive consensus control

来源：评论

学校读者我要写书评

暂无评论

A Learning Convolutional Neural Network Approach for Network Robustness Prediction

引用

IEEE TRANSACTIONS ON CYBERNETICS 2023年第7期53卷 4531-4544页

作者： Lou, Yang Wu, Ruizi Li, Junli Wang, Lin Li, Xiang Chen, Guanrong Sichuan Normal Univ Coll Comp Sci Chengdu 610066 Peoples R China Lingnan Univ Dept Comp & Decis Sci Hong Kong Peoples R China Shanghai Jiao Tong Univ Dept Automat Minist Educ Shanghai 200240 Peoples R China Shanghai Jiao Tong Univ Key Lab Syst Control & Informat Proc Minist Educ Shanghai 200240 Peoples R China Tongji Univ Shanghai Res Inst Intelligent Autonomous Syst Inst Complex Networks & Intelligent Syst Shanghai 201210 Peoples R China Tongji Univ Dept Control Sci & Engn Shanghai 200240 Peoples R China City Univ Hong Kong Dept Elect Engn Hong Kong Peoples R China

Network robustness is critical for various societal and industrial networks again malicious attacks. In particular, connectivity robustness and controllability robustness reflect how well a networked system can maintain its connectedness and controllability against destructive attacks, which can be quantified by a sequence of values that record the remaining connectivity and controllability of the network after a sequence of node- or edge-removal attacks. Traditionally, robustness is determined by attack simulations, which are computationally very time-consuming or even practically infeasible. In this paper, an improved method for network robustness prediction is developed based on learning feature representation using convolutional neural network (LFR-CNN). In this scheme, higher-dimensional network data are compressed to lower-dimensional representations, and then passed to a CNN to perform robustness prediction. Extensive experimental studies on both synthetic and real-world networks, both directed and undirected, demonstrate that 1) the proposed LFR-CNN performs better than other two state-of-the-art prediction methods, with significantly lower prediction errors;2) LFR-CNN is insensitive to the variation of the network size, which significantly extends its applicability;3) although LFR-CNN needs more time to perform feature learning, it can achieve accurate prediction faster than attack simulations;4) LFR-CNN not only can accurately predict network robustness, but also provides a good indicator for connectivity robustness, better than the classical spectral measures.

关键词： Robustness controllability complex networks Convolutional neural networks Optimization Neural networks Indexes complex network convolutional neural network (CNN) graph representation learning prediction robustness

来源：评论

学校读者我要写书评

暂无评论

Predicting epidemic threshold in complex networks by graph neural network

引用

CHAOS 2024年第6期34卷 063129-063129页

作者： Wang, Wu Li, Cong Qu, Bo Li, Xiang Fudan Univ Sch Informat Sci & Technol Dept Elect Engn Adapt Networks & Control Lab Shanghai 200433 Peoples R China HKCT Inst Higher Educ Inst Cyberspace Technol Hong Kong 999077 Peoples R China Tongji Univ Inst Complex Networks & Intelligent Syst Shanghai Res Inst Intelligent Autonomous Syst Shanghai 201210 Peoples R China

To achieve precision in predicting an epidemic threshold in complex networks, we have developed a novel threshold graph neural network (TGNN) that takes into account both the network topology and the spreading dynamical process, which together contribute to the epidemic threshold. The proposed TGNN could effectively and accurately predict the epidemic threshold in homogeneous networks, characterized by a small variance in the degree distribution, such as Erd & odblac;s-R & eacute;nyi random networks. Usability has also been validated when the range of the effective spreading rate is altered. Furthermore, extensive experiments in ER networks and scale-free networks validate the adaptability of the TGNN to different network topologies without the necessity for retaining. The adaptability of the TGNN is further validated in real-world networks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Analog control Circuit Designs for a Class of Continuous-Time Adaptive Fault-Tolerant control Systems

引用

IEEE TRANSACTIONS ON CYBERNETICS 2022年第6期52卷 4209-4220页

作者： Jin, Xiao-Zheng Che, Wei-Wei Wu, Zheng-Guang Wang, Hai Qilu Univ Technol Shandong Acad Sci Sch Comp Sci & Technol Jinan 250353 Peoples R China Qilu Univ Technol Shandong Comp Sci Ctr Natl Supercomp Ctr Jinan Jinan 250014 Peoples R China Qilu Univ Technol Shandong Prov Key Lab Comp Networks Jinan 250014 Peoples R China Qingdao Univ Inst Complex Sci Qingdao 266071 Peoples R China Zhejiang Univ Inst Cyber Syst & Control State Key Lab Ind Control Technol Hangzhou 310027 Peoples R China Murdoch Univ Discipline Engn & Energy Murdoch WA 6150 Australia

This article is concerned with the robust adaptive fault-tolerant control (FTC) circuit designs for a class of continuous-time disturbed systems. A circuit realization method is investigated to convert the robust adaptive FTC control schemes into analog control circuits. An adaptive compensation control scheme against state-dependent and partially bounded actuator faults and disturbances is first developed to demonstrate the approach clearly, then its equivalent control circuits are implemented by using the circuit theory. Compared with simulation results achieved by MATlab and professional circuit simulation software, the effectiveness of the proposed robust adaptive FTC circuits is validated by a rocket fairing system and a Chua's circuit system.

关键词： Actuators Circuit faults Adaptive systems Fault tolerance Fault tolerant systems Adaptation models Analog control circuits continuous-time disturbed systems hardware implementation robust adaptive fault-tolerant control (FTC) control schemes

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：