Pub Date : 2024-11-15DOI: 10.1016/j.ins.2024.121642
Peter Eades , Seokhee Hong , Giuseppe Liotta , Fabrizio Montecchiani , Martin Nöllenburg , Tommaso Piselli , Stephen Wismath
Motivated by the need for decision-making systems that avoid bias and discrimination, the concept of fairness recently gained traction in the broad field of artificial intelligence, stimulating new research also within the information visualization community. In this paper, we introduce a notion of fairness in network visualization, specifically for orthogonal and for straight-line drawings of graphs, two foundational paradigms in the field. We investigate the following research questions: (i) What is the price, in terms of global readability, of incorporating fairness constraints in graph drawings? (ii) How unfair is a graph drawing that does not optimize fairness as a primary objective? We present both theoretical and empirical results. In particular, we design and implement two optimization algorithms for multi-objective functions, one based on an ILP model for orthogonal drawings, and one based on gradient descent for straight-line drawings. In a nutshell, we experimentally show that it is possible to significantly increase the fairness of a drawing by paying a relatively small amount in terms of reduced global readability. Also, we present a use case in which we qualitatively evaluate our approach on a practical scenario.
{"title":"Introducing fairness in network visualization","authors":"Peter Eades , Seokhee Hong , Giuseppe Liotta , Fabrizio Montecchiani , Martin Nöllenburg , Tommaso Piselli , Stephen Wismath","doi":"10.1016/j.ins.2024.121642","DOIUrl":"10.1016/j.ins.2024.121642","url":null,"abstract":"<div><div>Motivated by the need for decision-making systems that avoid bias and discrimination, the concept of fairness recently gained traction in the broad field of artificial intelligence, stimulating new research also within the information visualization community. In this paper, we introduce a notion of fairness in network visualization, specifically for orthogonal and for straight-line drawings of graphs, two foundational paradigms in the field. We investigate the following research questions: (i) What is the price, in terms of global readability, of incorporating fairness constraints in graph drawings? (ii) How unfair is a graph drawing that does not optimize fairness as a primary objective? We present both theoretical and empirical results. In particular, we design and implement two optimization algorithms for multi-objective functions, one based on an ILP model for orthogonal drawings, and one based on gradient descent for straight-line drawings. In a nutshell, we experimentally show that it is possible to significantly increase the fairness of a drawing by paying a relatively small amount in terms of reduced global readability. Also, we present a use case in which we qualitatively evaluate our approach on a practical scenario.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"691 ","pages":"Article 121642"},"PeriodicalIF":8.1,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658130","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-15DOI: 10.1016/j.ins.2024.121646
Licheng Sun , Hongbin Ma , Zhentao Guo
In many real-world scenarios, tasks involve coordinating multiple agents, such as managing robot clusters, drone swarms, and autonomous vehicles. These tasks are commonly addressed using Multi-Agent Reinforcement Learning (MARL). However, existing MARL algorithms often lack foresight regarding the number and types of agents involved, requiring agents to generalize across various task configurations. This may lead to suboptimal performance due to underestimated action values and the selection of less effective joint policies. To address these challenges, we propose a novel multi-agent deep reinforcement learning framework, called multi-agent reinforcement learning framework based on adaptive grouping dynamic topological space (GDT). GDT utilizes a group mesh topology to interconnect the local action value functions of each agent, enabling effective coordination and knowledge sharing among agents. By computing three different interpretations of action value functions, GDT overcomes monotonicity constraints and derives more effective overall action value functions. Additionally, GDT groups agents with high similarity to facilitate parameter sharing, thereby enhancing knowledge transfer and generalization across different scenarios. Furthermore, GDT introduces a strategy regularization method for optimal exploration of multiple action spaces. This method assigns each agent an independent entropy temperature during exploration, enabling agents to efficiently explore potential actions and approximate total state values. Experimental results demonstrate that our approach, termed GDT, significantly outperforms state-of-the-art algorithms on Google Research Football (GRF) and the StarCraft Multi-Agent Challenge (SMAC). Particularly in SMAC tasks, GDT achieves a success rate of nearly 100% across almost all Hard Map and Super Hard Map scenarios. Additionally, we validate the effectiveness of our algorithm on Non-monotonic Matrix Games.
{"title":"GDT: Multi-agent reinforcement learning framework based on adaptive grouping dynamic topological space","authors":"Licheng Sun , Hongbin Ma , Zhentao Guo","doi":"10.1016/j.ins.2024.121646","DOIUrl":"10.1016/j.ins.2024.121646","url":null,"abstract":"<div><div>In many real-world scenarios, tasks involve coordinating multiple agents, such as managing robot clusters, drone swarms, and autonomous vehicles. These tasks are commonly addressed using Multi-Agent Reinforcement Learning (MARL). However, existing MARL algorithms often lack foresight regarding the number and types of agents involved, requiring agents to generalize across various task configurations. This may lead to suboptimal performance due to underestimated action values and the selection of less effective joint policies. To address these challenges, we propose a novel multi-agent deep reinforcement learning framework, called multi-agent reinforcement learning framework based on adaptive grouping dynamic topological space (GDT). GDT utilizes a group mesh topology to interconnect the local action value functions of each agent, enabling effective coordination and knowledge sharing among agents. By computing three different interpretations of action value functions, GDT overcomes monotonicity constraints and derives more effective overall action value functions. Additionally, GDT groups agents with high similarity to facilitate parameter sharing, thereby enhancing knowledge transfer and generalization across different scenarios. Furthermore, GDT introduces a strategy regularization method for optimal exploration of multiple action spaces. This method assigns each agent an independent entropy temperature during exploration, enabling agents to efficiently explore potential actions and approximate total state values. Experimental results demonstrate that our approach, termed GDT, significantly outperforms state-of-the-art algorithms on Google Research Football (GRF) and the StarCraft Multi-Agent Challenge (SMAC). Particularly in SMAC tasks, GDT achieves a success rate of nearly 100% across almost all Hard Map and Super Hard Map scenarios. Additionally, we validate the effectiveness of our algorithm on Non-monotonic Matrix Games.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"691 ","pages":"Article 121646"},"PeriodicalIF":8.1,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658131","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-15DOI: 10.1016/j.ins.2024.121647
Jie Mi , Huaiqin Wu , Jinde Cao
This article is concerned with the finite-time secure synchronization (FNTS) in mean square for stochastic complex networks (SCNs) with time-varying delayed coupling under deception attacks, where attack is described by a Bernoulli's stochastic variable, and is performed in the communication channel between the controller and the actuator. With the help of an auxiliary function, a new Halanay inequality is developed for continuous differential stochastic functions. By utilizing the Lyapunov functional gradient inequality with variable coefficients, a criterion about the finite-time stability in mean square is established for nonlinear stochastic systems under the designed two-step attenuation scheme. In order to reduce controller update consumption and communication waste, a two-step switching control mechanism consisting of an event-triggered control (ETC) and a time-varying gain state feedback control, is devised to achieve the FNTS objective. By Lyapunov stability theory, inequality analysis technique and the proposed finite-time stability criterion, the finite-time synchronization conditions are addressed in terms of linear matrix inequality (LMIs), and the bound of stochastic settling time (SST) is estimated explicitly. Finally, a practical application example is given to illustrate the effectiveness of the proposed control scheme, and to verify the correctness of the analytical results.
{"title":"Finite-time secure synchronization for stochastic complex networks with delayed coupling under deception attacks: A two-step switching control scheme","authors":"Jie Mi , Huaiqin Wu , Jinde Cao","doi":"10.1016/j.ins.2024.121647","DOIUrl":"10.1016/j.ins.2024.121647","url":null,"abstract":"<div><div>This article is concerned with the finite-time secure synchronization (FNTS) in mean square for stochastic complex networks (SCNs) with time-varying delayed coupling under deception attacks, where attack is described by a Bernoulli's stochastic variable, and is performed in the communication channel between the controller and the actuator. With the help of an auxiliary function, a new Halanay inequality is developed for continuous differential stochastic functions. By utilizing the Lyapunov functional gradient inequality with variable coefficients, a criterion about the finite-time stability in mean square is established for nonlinear stochastic systems under the designed two-step attenuation scheme. In order to reduce controller update consumption and communication waste, a two-step switching control mechanism consisting of an event-triggered control (ETC) and a time-varying gain state feedback control, is devised to achieve the FNTS objective. By Lyapunov stability theory, inequality analysis technique and the proposed finite-time stability criterion, the finite-time synchronization conditions are addressed in terms of linear matrix inequality (LMIs), and the bound of stochastic settling time (SST) is estimated explicitly. Finally, a practical application example is given to illustrate the effectiveness of the proposed control scheme, and to verify the correctness of the analytical results.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"691 ","pages":"Article 121647"},"PeriodicalIF":8.1,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658128","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-15DOI: 10.1016/j.ins.2024.121649
Zengyou He , Xiaolei Li , Lianyu Hu , Mudi Jiang , Yan Liu
The detection of communities from a graph is a key issue in network science and graph data mining. However, existing community detection algorithms can always partition a given network/graph into different communities/subgraphs, even when no community structure exists. Obviously, it will lead to fruitless efforts and erroneous conclusions if we conduct the community detection procedure on a network without a community structure. Hence, prior to community detection, it is a must to test whether the community structure is present in the target network. Unfortunately, the community structure testing issue is still not revolved and existing solutions have some limitations. Therefore, we present a new test, which is called FCN (Frequent Common Neighbor) test to tackle the community structure testing problem. In FCN test, the number of FCN sets is employed as the test statistic, which will approximately follows a Poisson distribution when the support threshold is sufficiently large under the null hypothesis that the graph is generated according to the Erdős-Rényi model. We compare the proposed FCN test with existing community structure testing methods on both real networks and simulated networks. The experimental results demonstrate the effectiveness and advantage of our method.
{"title":"Community structure testing by counting frequent common neighbor sets","authors":"Zengyou He , Xiaolei Li , Lianyu Hu , Mudi Jiang , Yan Liu","doi":"10.1016/j.ins.2024.121649","DOIUrl":"10.1016/j.ins.2024.121649","url":null,"abstract":"<div><div>The detection of communities from a graph is a key issue in network science and graph data mining. However, existing community detection algorithms can always partition a given network/graph into different communities/subgraphs, even when no community structure exists. Obviously, it will lead to fruitless efforts and erroneous conclusions if we conduct the community detection procedure on a network without a community structure. Hence, prior to community detection, it is a must to test whether the community structure is present in the target network. Unfortunately, the community structure testing issue is still not revolved and existing solutions have some limitations. Therefore, we present a new test, which is called FCN (Frequent Common Neighbor) test to tackle the community structure testing problem. In FCN test, the number of FCN sets is employed as the test statistic, which will approximately follows a Poisson distribution when the support threshold is sufficiently large under the null hypothesis that the graph is generated according to the Erdős-Rényi model. We compare the proposed FCN test with existing community structure testing methods on both real networks and simulated networks. The experimental results demonstrate the effectiveness and advantage of our method.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"691 ","pages":"Article 121649"},"PeriodicalIF":8.1,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658127","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-15DOI: 10.1016/j.ins.2024.121644
Kecan Cai , Hongyun Zhang , Miao Li , Duoqian Miao
Efficiency is crucial in deep learning tasks and has garnered significant attention in green deep learning research field. However, existing methods often sacrifice efficiency for slight accuracy improvement, requiring extensive computational resources. This paper proposes an adaptive granular data compression and interval granulation method to improve classification efficiency without compromising accuracy. The approach comprises two main components: Adaptive Granular Data Compression (AG), and Interval Granulation (IG). Specifically, AG employs principle of justifiable granularity for adaptive generating granular data. AG enables the extraction of abstract granular subset representations from the original dataset, capturing essential features and thereby reducing computational complexity. The quality of the generated granular data is evaluated using coverage and specificity criteria, which are standard metrics in evaluating information granules. Furthermore, the design of IG performs AG operation on the input data at regular intervals during the training process. The multiple regular granulation operations during the training process increase the diversity of samples and help the model achieve better training. It is noteworthy that the proposed method can be extended to any convolution-based and attention-based classification neural network. Extensive experiments conducted on benchmark datasets demonstrate that the proposed method significantly enhances the classification efficiency without compromising accuracy.
效率在深度学习任务中至关重要，在绿色深度学习研究领域备受关注。然而，现有的方法往往牺牲效率来换取微小的准确率提升，这需要大量的计算资源。本文提出了一种自适应粒度数据压缩和区间粒度化方法，以在不影响准确性的前提下提高分类效率。该方法由两个主要部分组成：自适应粒度数据压缩（AG）和间隔粒化（IG）。具体来说，AG 采用合理粒度原则自适应生成粒度数据。AG 可以从原始数据集中提取抽象的粒度子集表示，捕捉基本特征，从而降低计算复杂度。生成的粒度数据的质量使用覆盖率和特异性标准进行评估，这两个标准是评估信息粒度的标准指标。此外，IG 的设计在训练过程中定期对输入数据执行 AG 操作。训练过程中的多次定时颗粒化操作增加了样本的多样性，有助于模型实现更好的训练效果。值得注意的是，所提出的方法可以扩展到任何基于卷积和注意力的分类神经网络。在基准数据集上进行的大量实验证明，所提出的方法能在不影响准确性的前提下显著提高分类效率。
{"title":"Adaptive granular data compression and interval granulation for efficient classification","authors":"Kecan Cai , Hongyun Zhang , Miao Li , Duoqian Miao","doi":"10.1016/j.ins.2024.121644","DOIUrl":"10.1016/j.ins.2024.121644","url":null,"abstract":"<div><div>Efficiency is crucial in deep learning tasks and has garnered significant attention in green deep learning research field. However, existing methods often sacrifice efficiency for slight accuracy improvement, requiring extensive computational resources. This paper proposes an adaptive granular data compression and interval granulation method to improve classification efficiency without compromising accuracy. The approach comprises two main components: Adaptive Granular Data Compression (AG), and Interval Granulation (IG). Specifically, AG employs principle of justifiable granularity for adaptive generating granular data. AG enables the extraction of abstract granular subset representations from the original dataset, capturing essential features and thereby reducing computational complexity. The quality of the generated granular data is evaluated using coverage and specificity criteria, which are standard metrics in evaluating information granules. Furthermore, the design of IG performs AG operation on the input data at regular intervals during the training process. The multiple regular granulation operations during the training process increase the diversity of samples and help the model achieve better training. It is noteworthy that the proposed method can be extended to any convolution-based and attention-based classification neural network. Extensive experiments conducted on benchmark datasets demonstrate that the proposed method significantly enhances the classification efficiency without compromising accuracy.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"691 ","pages":"Article 121644"},"PeriodicalIF":8.1,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658129","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-14DOI: 10.1016/j.ins.2024.121641
Yi Zhang , Yunfan Lu , Fengxia Liu , Cheng Li , Zixian Gong , Zhe Hu , Qun Xu
Federated learning constitutes a paradigm in distributed machine learning, wherein model training unfolds through the exchange of intermediary results between a central server and federated clients. Given its decentralized nature, conventional machine learning algorithms find limited applicability in the context of federated learning models. Hence, the alternating direction method of multipliers (ADMM), tailored for distributed optimization, is leveraged for this purpose. However, despite the considerable promise of the ADMM algorithm in federated learning, it faces challenges related to computational efficiency, communication efficiency, and data security. In response to these challenges, this study proposes the privacy-preserving and communication-efficient stochastic ADMM (PPCESADMM) algorithm that enhances the computational efficiency through the stochastic optimization method, reduces communication costs through sparse communication method, and ensures the security of federated clients' data via the homomorphic encryption method. Theoretical analyses confirm the convergence of the PPCESADMM algorithm under mild conditions and establish its convergence rate as . Experiments illustrate the superior performance of our algorithm in communication cost compared to ADMM and CEADMM algorithms, achieving reductions of 65.10% and 44.32%, respectively. Furthermore, our method surpasses classical federated learning algorithms such as FedAvg, FedAvgM, and SCAFFOLD in terms of algorithmic convergence, achieving superior convergence precision within predefined training epochs. Finally, our algorithm converges to the same results as those obtained without using homomorphic encryption, albeit at the cost of increased computation time.
{"title":"Privacy-preserving and communication-efficient stochastic alternating direction method of multipliers for federated learning","authors":"Yi Zhang , Yunfan Lu , Fengxia Liu , Cheng Li , Zixian Gong , Zhe Hu , Qun Xu","doi":"10.1016/j.ins.2024.121641","DOIUrl":"10.1016/j.ins.2024.121641","url":null,"abstract":"<div><div>Federated learning constitutes a paradigm in distributed machine learning, wherein model training unfolds through the exchange of intermediary results between a central server and federated clients. Given its decentralized nature, conventional machine learning algorithms find limited applicability in the context of federated learning models. Hence, the alternating direction method of multipliers (ADMM), tailored for distributed optimization, is leveraged for this purpose. However, despite the considerable promise of the ADMM algorithm in federated learning, it faces challenges related to computational efficiency, communication efficiency, and data security. In response to these challenges, this study proposes the privacy-preserving and communication-efficient stochastic ADMM (PPCESADMM) algorithm that enhances the computational efficiency through the stochastic optimization method, reduces communication costs through sparse communication method, and ensures the security of federated clients' data via the homomorphic encryption method. Theoretical analyses confirm the convergence of the PPCESADMM algorithm under mild conditions and establish its convergence rate as <span><math><mi>O</mi><mo>(</mo><mn>1</mn><mo>/</mo><msqrt><mrow><mi>T</mi></mrow></msqrt><mo>)</mo></math></span>. Experiments illustrate the superior performance of our algorithm in communication cost compared to ADMM and CEADMM algorithms, achieving reductions of 65.10% and 44.32%, respectively. Furthermore, our method surpasses classical federated learning algorithms such as FedAvg, FedAvgM, and SCAFFOLD in terms of algorithmic convergence, achieving superior convergence precision within predefined training epochs. Finally, our algorithm converges to the same results as those obtained without using homomorphic encryption, albeit at the cost of increased computation time.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"691 ","pages":"Article 121641"},"PeriodicalIF":8.1,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658126","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-13DOI: 10.1016/j.ins.2024.121639
Zhifei Li , Lifan Chen , Yue Jian , Han Wang , Yue Zhao , Miao Zhang , Kui Xiao , Yan Zhang , Honglian Deng , Xiaoju Hou
Knowledge graph completion intends to infer information within knowledge graphs, thereby bolstering the functionality of knowledge-driven applications. Recently, there has been a significant increase in the utilization of graph convolutional networks (GCNs) for knowledge graph completion. These GCN-based models primarily focus on aggregating information from neighboring entities and relations. Nonetheless, a fundamental question arises: is it beneficial to consider all neighbor information, and should some neighbor features be separated? We tackle this issue and present an adaptive graph convolutional network (AdaGCN) for knowledge graph completion, which can adaptively aggregate or separate neighbor information for knowledge embedding learning. Specifically, AdaGCN utilizes the adaptive message-passing mechanism to determine the importance of each relation, allocating weights to neighbor entity embeddings. This adaptive approach facilitates the propagation of valuable information while effectively separating less relevant or unnecessary details. Experimental results demonstrate that AdaGCN can efficiently acquire the embeddings of various triplets within knowledge graphs, and it achieves competitive performance compared to SOTA models on six datasets for the tasks of knowledge graph completion.
知识图谱补全旨在推断知识图谱中的信息，从而增强知识驱动型应用的功能。最近，利用图卷积网络（GCN）完成知识图谱的情况显著增加。这些基于 GCN 的模型主要侧重于聚合相邻实体和关系的信息。然而，一个基本问题随之而来：考虑所有相邻信息是否有益，是否应该分离某些相邻特征？针对这一问题，我们提出了一种用于知识图谱补全的自适应图卷积网络（AdaGCN），它可以自适应地聚合或分离邻居信息，从而实现知识嵌入学习。具体来说，AdaGCN 利用自适应信息传递机制来确定每种关系的重要性，并为相邻实体嵌入分配权重。这种自适应方法有利于传播有价值的信息，同时有效分离相关性较低或不必要的细节。实验结果表明，AdaGCN 可以高效地获取知识图谱中各种三元组的嵌入信息，并且在六个数据集的知识图谱补全任务中取得了与 SOTA 模型相比具有竞争力的性能。
{"title":"Aggregation or separation? Adaptive embedding message passing for knowledge graph completion","authors":"Zhifei Li , Lifan Chen , Yue Jian , Han Wang , Yue Zhao , Miao Zhang , Kui Xiao , Yan Zhang , Honglian Deng , Xiaoju Hou","doi":"10.1016/j.ins.2024.121639","DOIUrl":"10.1016/j.ins.2024.121639","url":null,"abstract":"<div><div>Knowledge graph completion intends to infer information within knowledge graphs, thereby bolstering the functionality of knowledge-driven applications. Recently, there has been a significant increase in the utilization of graph convolutional networks (GCNs) for knowledge graph completion. These GCN-based models primarily focus on aggregating information from neighboring entities and relations. Nonetheless, a fundamental question arises: is it beneficial to consider all neighbor information, and should some neighbor features be separated? We tackle this issue and present an adaptive graph convolutional network (AdaGCN) for knowledge graph completion, which can adaptively aggregate or separate neighbor information for knowledge embedding learning. Specifically, AdaGCN utilizes the adaptive message-passing mechanism to determine the importance of each relation, allocating weights to neighbor entity embeddings. This adaptive approach facilitates the propagation of valuable information while effectively separating less relevant or unnecessary details. Experimental results demonstrate that AdaGCN can efficiently acquire the embeddings of various triplets within knowledge graphs, and it achieves competitive performance compared to SOTA models on six datasets for the tasks of knowledge graph completion.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"691 ","pages":"Article 121639"},"PeriodicalIF":8.1,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658123","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-12DOI: 10.1016/j.ins.2024.121626
Xinggui Zhao, Bo Meng, Zhen Wang
The dynamic quantization of event-triggered (ET) adaptive sliding mode control (SM, SMC) for networked control systems (NCS) under false data injection attack (FDIA) is considered in this article. To begin with, to reduce the network transmission burden, dynamic quantizers are used to quantize the states and the input on the channels from the plant to the ET mechanism and from the controller to the plant, respectively. Secondly, the dynamic ET mechanism employs quantized state error, and the existence of the minimum inter-event time demonstrates that the system does not experience the Zeno phenomenon. Thirdly, this paper uses the adaptive parameter to estimate the unknown upper bound of the attack mode. In addition, the range of values for the adaptive gain of the SMC is derived by combining with the Lyapunov stability theory. On the last, the comparative simulation results of different methods for numerical examples are given to verify the superiority of the method proposed in this paper.
本文研究了在虚假数据注入攻击（FDIA）下网络控制系统（NCS）的事件触发（ET）自适应滑模控制（SM，SMC）的动态量化问题。首先，为了减少网络传输负担，本文使用动态量化器分别量化从工厂到 ET 机制以及从控制器到工厂的通道上的状态和输入。其次，动态 ET 机制采用量化状态误差，最小事件间时间的存在表明系统不会出现芝诺现象。第三，本文利用自适应参数估计攻击模式的未知上限。此外，本文还结合李雅普诺夫稳定性理论，得出了 SMC 自适应增益的取值范围。最后，本文给出了不同方法的数值实例仿真结果对比，以验证本文所提方法的优越性。
{"title":"Dynamic quantization of event-triggered adaptive sliding mode control for networked control systems under false data injection attack","authors":"Xinggui Zhao, Bo Meng, Zhen Wang","doi":"10.1016/j.ins.2024.121626","DOIUrl":"10.1016/j.ins.2024.121626","url":null,"abstract":"<div><div>The dynamic quantization of event-triggered (ET) adaptive sliding mode control (SM, SMC) for networked control systems (NCS) under false data injection attack (FDIA) is considered in this article. To begin with, to reduce the network transmission burden, dynamic quantizers are used to quantize the states and the input on the channels from the plant to the ET mechanism and from the controller to the plant, respectively. Secondly, the dynamic ET mechanism employs quantized state error, and the existence of the minimum inter-event time demonstrates that the system does not experience the Zeno phenomenon. Thirdly, this paper uses the adaptive parameter to estimate the unknown upper bound of the attack mode. In addition, the range of values for the adaptive gain of the SMC is derived by combining with the Lyapunov stability theory. On the last, the comparative simulation results of different methods for numerical examples are given to verify the superiority of the method proposed in this paper.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"691 ","pages":"Article 121626"},"PeriodicalIF":8.1,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658099","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-09DOI: 10.1016/j.ins.2024.121624
Yiling Zeng , Chuanfeng Jian , Chunyao Song , Tingjian Ge , Yuhan Li , Yuqing Zhou
Heterogeneous graph streams represent data interactions in real-world applications and are characterized by dynamic and heterogeneous properties including varying node labels, edge labels and edge weights. The mining of graph streams is critical in fields such as network security, social network analysis, and traffic control. However, the sheer volume and high dynamics of graph streams pose significant challenges for efficient storage and accurate query analysis. To address these challenges, we propose LSketch, a novel sketch technique designed for heterogeneous graph streams. Unlike traditional methods, LSketch effectively preserves the diverse label information inherent in these streams, enhancing the expressive ability of sketches. Furthermore, as graph streams evolve over time, some edges may become outdated and lose their relevance. LSketch incorporates a sliding window model that eliminates expired edges, ensuring that the analysis remains focused on the most current and relevant data automatically. LSketch operates with sub-linear storage space and supports both structure-based and time-sensitive queries with high accuracy. We perform extensive experiments over four real datasets, demonstrating that LSketch outperforms state-of-the-art methods in terms of query accuracy and time efficiency.
{"title":"LSketch: A label-enabled graph stream sketch toward time-sensitive queries","authors":"Yiling Zeng , Chuanfeng Jian , Chunyao Song , Tingjian Ge , Yuhan Li , Yuqing Zhou","doi":"10.1016/j.ins.2024.121624","DOIUrl":"10.1016/j.ins.2024.121624","url":null,"abstract":"<div><div>Heterogeneous graph streams represent data interactions in real-world applications and are characterized by dynamic and heterogeneous properties including varying node labels, edge labels and edge weights. The mining of graph streams is critical in fields such as network security, social network analysis, and traffic control. However, the sheer volume and high dynamics of graph streams pose significant challenges for efficient storage and accurate query analysis. To address these challenges, we propose LSketch, a novel sketch technique designed for heterogeneous graph streams. Unlike traditional methods, LSketch effectively preserves the diverse label information inherent in these streams, enhancing the expressive ability of sketches. Furthermore, as graph streams evolve over time, some edges may become outdated and lose their relevance. LSketch incorporates a sliding window model that eliminates expired edges, ensuring that the analysis remains focused on the most current and relevant data automatically. LSketch operates with sub-linear storage space and supports both structure-based and time-sensitive queries with high accuracy. We perform extensive experiments over four real datasets, demonstrating that LSketch outperforms state-of-the-art methods in terms of query accuracy and time efficiency.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"691 ","pages":"Article 121624"},"PeriodicalIF":8.1,"publicationDate":"2024-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658024","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-08DOI: 10.1016/j.ins.2024.121628
Lei Meng , Guiqiong Xu , Chen Dong , Shoujin Wang
The rapid development of online social networks has greatly facilitated the dissemination and sharing of information. Effectively guiding the propagation of information to specific target groups is a significant and challenging research issue, which can be formulated as the target propagation problem. Most existing studies, however, focus on traditional information propagation methods, treating all users in the network as target audiences, which results in low efficiency and high costs. To address this issue, we propose a novel information propagation model that incorporates adaptive guidance and incentive strategies, called the model, to simulate the target spreading process in online social networks. Our model is designed to enhance both global communication capabilities and information transmission efficiency by introducing a mutual influence score that quantifies the interaction between target and non-target users. Based on this, the model adaptively guides and incentivizes non-target users to disseminate information specifically to target user groups. We conducted several groups of experiments on nine real-world social networks, assessing scenarios with both single and multiple target groups. Experimental results demonstrate that the model outperforms existing methods in terms of target influence range and the effectiveness of information spreading, thereby offering valuable insights for practical applications.
{"title":"Modeling information propagation for target user groups in online social networks based on guidance and incentive strategies","authors":"Lei Meng , Guiqiong Xu , Chen Dong , Shoujin Wang","doi":"10.1016/j.ins.2024.121628","DOIUrl":"10.1016/j.ins.2024.121628","url":null,"abstract":"<div><div>The rapid development of online social networks has greatly facilitated the dissemination and sharing of information. Effectively guiding the propagation of information to specific target groups is a significant and challenging research issue, which can be formulated as the <em>target propagation</em> problem. Most existing studies, however, focus on traditional information propagation methods, treating all users in the network as target audiences, which results in low efficiency and high costs. To address this issue, we propose a novel information propagation model that incorporates adaptive guidance and incentive strategies, called the <span><math><mi>S</mi><mi>I</mi><msub><mrow><mi>I</mi></mrow><mrow><mi>i</mi><mi>n</mi></mrow></msub><msub><mrow><mi>R</mi></mrow><mrow><mi>g</mi><mi>u</mi></mrow></msub></math></span> model, to simulate the target spreading process in online social networks. Our model is designed to enhance both global communication capabilities and information transmission efficiency by introducing a mutual influence score that quantifies the interaction between target and non-target users. Based on this, the <span><math><mi>S</mi><mi>I</mi><msub><mrow><mi>I</mi></mrow><mrow><mi>i</mi><mi>n</mi></mrow></msub><msub><mrow><mi>R</mi></mrow><mrow><mi>g</mi><mi>u</mi></mrow></msub></math></span> model adaptively guides and incentivizes non-target users to disseminate information specifically to target user groups. We conducted several groups of experiments on nine real-world social networks, assessing scenarios with both single and multiple target groups. Experimental results demonstrate that the <span><math><mi>S</mi><mi>I</mi><msub><mrow><mi>I</mi></mrow><mrow><mi>i</mi><mi>n</mi></mrow></msub><msub><mrow><mi>R</mi></mrow><mrow><mi>g</mi><mi>u</mi></mrow></msub></math></span> model outperforms existing methods in terms of target influence range and the effectiveness of information spreading, thereby offering valuable insights for practical applications.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"691 ","pages":"Article 121628"},"PeriodicalIF":8.1,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658133","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号