检索结果-内蒙古大学图书馆

61st ACM/IEEE Design Automation Conference, DAC 2024

作者： Yu, Hui Zhang, Yu He, Ligang He, Donghao Li, Qikun Zhao, Jin Liao, Xiaofei Jin, Hai Gu, Lin Liu, Haikun National Engineering Research Center for Big Data Technology and System Service Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology China Department of Computer Science University of Warwick United Kingdom

ISBN: (纸本)9798400706011

Asynchronous Graph Neural Network (AGNN) has attracted much research attention because it enables faster convergence speed than the synchronous GNN. However, existing software/hardware solutions suffer from redundant computation overhead and excessive off-chip communications for AGNN due to irregular state propagations along the dependency chains between vertices. This paper proposes a chain-driven asynchronous accelerator, CDA-GNN, for efficient AGNN inference. Specifically, CDA-GNN proposes a chain-driven asynchronous execution approach into novel accelerator design to regularize the vertex state propagations for fewer redundant computations and off-chip communications and also designs a chain-aware data caching method to improve data locality for AGNN. We have implemented and evaluated CDA-GNN on a Xilinx Alveo U280 FPGA card. Compared with the cutting-edge software solutions (i.e., Dorylus and AMP) and hardware solutions (i.e., BlockGNN and FlowGNN), CDA-GNN improves the performance of AGNN inference by an average of 1,173x, 182.4x, 10.2x, and 7.9x and saves energy by 2,241x, 242.2x, 12.4x, and 8.9x, respectively. © 2024 Copyright is held by the owner/author(s). Publication rights licensed to ACM.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Incremental Learning Algorithms for Broad Learning System with Node and Input Addition

Incremental Learning Algorithms for Broad Learning System wi...

引用

2024 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2024

作者： Chen, Guang-Ze Jin, Junwei Sun, Hai-Wei Philip Chen, C.L. Faculty of Science and Technology University of Macau Department of Mathematics 99999 China School of Artificial Intelligence and Big Data Henan University of Technology Zhengzhou450001 China School of Computer Science and Engineering South China University of Technology and Pazhou Lab Guangzhou510641 510335 China

ISBN: (纸本)9781665410205

The Broad Learning System (BLS) has been established as an effective flat network alternative to Deep Neural Networks (DNNs), delivering high efficiency while achieving competitive accuracy. Despite its advantages, the incremental learning methods of BLS face challenges in stability and computation when expanding with new nodes or input. We introduce two novel incremental learning algorithms based on factorization updates for BLS that optimize node and input additions to overcome these limitations. Our node addition algorithm utilizes QR decomposition and Cholesky factorization, using the update of the Cholesky factor instead of pseudo-inverse computations. For input addition, we propose an iterative Cholesky factor update algorithm. Our algorithms demonstrate not only faster computation compared to the existing BLS but also improved testing accuracy on the MNIST or Fashion-MNIST dataset. This work presents a significant step forward in the practical application and scalability of BLS in various data-dense environments. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Effective and Imperceptible Adversarial Textual Attack Via Multi-objectivization

引用

ACM Transactions on Evolutionary Learning and Optimization 2024年第3期4卷 1-23页

作者： Liu, Shengcai Lu, Ning Hong, Wenjing Qian, Chao Tang, Ke Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen College of Engineering South Tower 1088 Xueyuan Avenue Guangdong Shenzhen518055 China Department of Computer Science and Engineering Hong Kong University of Science and Technology Hong Kong999077 Hong Kong National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen 10F Zhizhen Building Canghai Campus 3688 Nanhai Avenue Guangdong Shenzhen China National Key Laboratory for Novel Software Technology School of Artificial Intelligence Nanjing University Nanjing International College Area Xianlin Campus 163 Xianlin Avenue Jiangsu Nanjing China

The field of adversarial textual attack has significantly grown over the past few years, where the commonly considered objective is to craft adversarial examples (AEs) that can successfully fool the target model. However, the imperceptibility of attacks, which is also essential for practical attackers, is often left out by previous studies. In consequence, the crafted AEs tend to have obvious structural and semantic differences from the original human-written text, making them easily perceptible. In this work, we advocate leveraging multi-objectivization to address such an issue. Specifically, we reformulate the problem of crafting AEs as a multi-objective optimization problem, where the attack imperceptibility is considered as an auxiliary objective. Then, we propose a simple yet effective evolutionary algorithm, dubbed HydraText, to solve this problem. HydraText can be effectively applied to both score-based and decision-based attack settings. Exhaustive experiments involving 44,237 instances demonstrate that HydraText consistently achieves competitive attack success rates and better attack imperceptibility than the recently proposed attack approaches. A human evaluation study also shows that the AEs crafted by HydraText are more indistinguishable from human-written text. Finally, these AEs exhibit good transferability and can bring notable robustness improvement to the target model by adversarial training. © 2024 Copyright held by the owner/author(s).

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

Two-Stage Guided Constraint Differential Evolution Algorithm

Journal of Network Intelligence

引用

Journal of Network intelligence 2023年第4期8卷 1109-1133页

作者： Sung, Tien-Wen Liang, Qiaoxin Hong, Chuanbo Huang, Zeming Li, Wei Nguyen, Trinh-Dong College of Computer Science and Mathematics Fujian Provincial Key Laboratory of Big Data Mining and Applications Fujian University of Technology Fujian Fuzhou350118 China School of Electronic Electrical Engineering and Physics Fujian Provincial Key Laboratory of Big Data Mining and Applications Fujian University of Technology Fujian Fuzhou350118 China Department of Software Engineering University of Information Technology Ho Chi Minh City Viet Nam

The development of population intelligence has shown a great trend of a linear surge in recent years, and a large number of intelligent algorithms inspired by biology have been studied. Among them, differential evolution (DE) algorithms have attracted the attention of many researchers because of their simplicity, high efficiency, powerful search capability, etc. DE has strong exploitability and can approach the global optimum faster, but it cannot avoid the phenomenon of search stagnation and premature convergence. To address the drawbacks of DE, the two-stage bootstrap constrained differential evolution algorithm (TSGCDE) is proposed in this paper. In TSGCDE, instead of the traditional one population, the population is divided into two as a way to increase the diversity of the population. Divided into two stages, each stage searches for optimal solutions in different directions in different populations, and the optimal solutions and the bootstrap solutions generated by the populations in the two stages are bounded by each other’s bootstrap, which makes the DE have a more powerful optimal-seeking ability and easier to jump out of the local optimum. To validate the performance of TSGCDE, we conducted experiments on CEC2013 and CEC2014 benchmarks with some well-known algorithm variants, and the experimental results show that our proposed algorithm is competitive. © 2023, Taiwan Ubiquitous Information CO LTD. All rights reserved.

关键词： Swarm intelligence

来源：评论

学校读者我要写书评

暂无评论

How Graph Neural Networks Learn: Lessons from Training Dynamics 41

How Graph Neural Networks Learn: Lessons from Training Dynam...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Yang, Chenxiao Wu, Qitian Wipf, David Sun, Ruoyu Yan, Junchi School of Artificial Intelligence Department of Computer Science and Engineering MoE Lab of AI Shanghai Jiao Tong University China Amazon Web Services United States School of Data Science The Chinese University of Hong Kong Shenzhen China Shenzhen International Center for Industrial and Applied Mathematics Shenzhen Research Institute of Big Data China

A long-standing goal in deep learning has been to characterize the learning behavior of black-box models in a more interpretable manner. For graph neural networks (GNNs), considerable advances have been made in formalizing what functions they can represent, but whether GNNs will learn desired functions during the optimization process remains less clear. To fill this gap, we study their training dynamics in function space. In particular, we find that the gradient descent optimization of GNNs implicitly leverages the graph structure to update the learned function, as can be quantified by a phenomenon which we call kernel-graph alignment. We provide theoretical explanations for the emergence of this phenomenon in the overparameterized regime and empirically validate it on real-world GNNs. This finding offers new interpretable insights into when and why the learned GNN functions generalize, highlighting their limitations in heterophilic graphs. Practically, we propose a parameter-free algorithm that directly uses a sparse matrix (i.e. graph adjacency) to update the learned function. We demonstrate that this embarrassingly simple approach can be as effective as GNNs while being orders-of-magnitude faster. Copyright 2024 by the author(s)

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Privacy Enhanced Mobile User Authentication Method Using Motion Sensors

引用

computer Modeling in engineering & sciences 2024年第3期138卷 3013-3032页

作者： Chunlin Xiong Zhengqiu Weng Jia Liu Liang Gu Fayez Alqahtani Amr Gafar Pradip Kumar Sharma Shenzhen Institutes of Advanced Technology Chinese Academy of SciencesShenzhen518052China Sangfor Technologies Inc. Shenzhen518055China School of Data Science and Artificial Intelligence Wenzhou University of TechnologyWenzhou325035China College of Computer Science&Technology Zhejiang University of TechnologyHangzhou310023China Software Engineering Department College of Computer and Information SciencesKing Saud UniversityRiyadh12372Saudi Arabia Math&Computer Science Department Faculty of ScienceMenofia UniversityShebin El-KomEgypt Computing Science Department University of AberdeenAberdeenUK

With the development of hardware devices and the upgrading of smartphones,a large number of users save privacy-related information in mobile devices,mainly smartphones,which puts forward higher demands on the protection of mobile users’privacy *** present,mobile user authenticationmethods based on humancomputer interaction have been extensively studied due to their advantages of high precision and non-perception,but there are still shortcomings such as low data collection efficiency,untrustworthy participating nodes,and lack of *** this end,this paper proposes a privacy-enhanced mobile user authentication method with motion sensors,which mainly includes:(1)Construct a smart contract-based private chain and federated learning to improve the data collection efficiency of mobile user authentication,reduce the probability of the model being bypassed by attackers,and reduce the overhead of data centralized processing and the risk of privacy leakage;(2)Use certificateless encryption to realize the authentication of the device to ensure the credibility of the client nodes participating in the calculation;(3)Combine Variational Mode Decomposition(VMD)and Long Short-TermMemory(LSTM)to analyze and model the motion sensor data of mobile devices to improve the accuracy of model *** experimental results on the real environment dataset of 1513 people show that themethod proposed in this paper can effectively resist poisoning attacks while ensuring the accuracy and efficiency of mobile user authentication.

关键词： Mobile authentication blockchain federated learning smart contract certificateless encryption VMD LSTM

来源：评论

学校读者我要写书评

暂无评论

Quantus: An Explainable AI Toolkit for Responsible Evaluation of Neural Network Explanations and Beyond

引用

Journal of Machine Learning Research 2023年第1期24卷 1-11页

作者： Hedström, Anna Weber, Leander Bareeva, Dilyara Krakowczyk, Daniel Motzkus, Franz Samek, Wojciech Lapuschkin, Sebastian Höhne, Marina M.-C. Understandable Machine Intelligence Lab TU Berlin Berlin10587 Germany Department of Electrical Engineering and Computer Science TU Berlin Berlin10587 Germany Department of Artificial Intelligence Fraunhofer Heinrich-Hertz-Institute Berlin10587 Germany Department of Computer Science University of Potsdam Potsdam14476 Germany BIFOLD – Berlin Institute for the Foundations of Learning and Data Berlin10587 Germany

The evaluation of explanation methods is a research topic that has not yet been explored deeply, however, since explainability is supposed to strengthen trust in artificial intelligence, it is necessary to systematically review and compare explanation methods in order to confirm their correctness. Until now, no tool with focus on XAI evaluation exists that exhaustively and speedily allows researchers to evaluate the performance of explanations of neural network predictions. To increase transparency and reproducibility in the field, we therefore built Quantus—a comprehensive, evaluation toolkit in Python that includes a growing, wellorganised collection of evaluation metrics and tutorials for evaluating explainable methods. The toolkit has been thoroughly tested and is available under an open-source license on PyPi (or on https://***/understandable-machine-intelligence-lab/Quantus/). © 2023 Anna Hedström, Leander Weber, Dilyara Bareeva, Daniel Krakowczyk, Franz Motzkus, Wojciech Samek, Sebastian Lapuschkin, and Marina M.-C. Höhne.

关键词： Open systems

来源：评论

学校读者我要写书评

暂无评论

Evolutionary Multiobjective Feature Selection Assisted by Unselected Features 13

Evolutionary Multiobjective Feature Selection Assisted by Un...

引用

13th IEEE Congress on Evolutionary Computation, CEC 2024

作者： Duan, Xuan Liu, Songbai Ji, Junkai Li, Lingjie Lin, Qiuzhen Tan, Kay Chen College of Computer Science and Software Engineering Shenzhen University Shenzhen China Shenzhen University National Engineering Laboratory for Big Data System Computing Technology Shenzhen China Shenzhen China The Hong Kong Polytechnic University Department of Computing Hong Kong

ISBN: (纸本)9798350308365

To enhance the generalization of multi-objective feature selection (MOFS) in classification, this paper proposes an evolutionary multitasking algorithm, diverging from previous approaches that exclusively target selected features. The algorithm integrates information from both selected and unselected features, introducing a novel objective to minimize the accuracy of unselected features. This objective, combined with the goal of minimizing classification errors for selected features, forms an auxiliary MOFS task. The paper presents a dual-population evolutionary multitasking framework that synergizes the main MOFS task with the auxiliary task. A knowledge transfer mechanism, based on accuracy preferences, seamlessly shares insights from the auxiliary to the main task, aiming to identify improved Pareto feature subsets. Empirical results demonstrate the superior performance of several state-of-the-art multi-objective algorithms within this framework, highlighting significant improvements across diverse datasets. © 2024 IEEE.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Dynamic Topology Management In Ad-Hoc Networks For Improved Performance 2

Dynamic Topology Management In Ad-Hoc Networks For Improved ...

引用

2nd IEEE International Conference on Advances in Information Technology, ICAIT 2024

作者： Anakath, A.S. Kannadasan, R. Margarat, G. Simi Pandian, A. Pasumpon Antony Sibiya Varghese, V. Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Tamil Nadu Chennai India Department of Software Systems School of Computer Science and Engineering Vellore Institute of Technology VIT University Tamil Nadu Vellore India Department of Computer Science and Engineering New Prince Shri Bhavani College of Engineering and Technology Tamil Nadu Chennai India Department of Computer Science and Engineering CARE College of Engineering Trichy India Department of Artificial Intelligence and Data Science GRT Institute of Engineering and Technology Tamil Nadu Tiruttani India

ISBN: (纸本)9798350383867

Optimization in ad hoc networks is a highly specialized task because the structure of the network is loosely formed, and each node is independently responsible for its operation. In order to obtain a better result from this solely formed networks this report offer extensive Dynamic Topology Management (DTM) method. Dynamic routing protocols lager lines, dividend balancing, Life-saving administration, guarantee of reliability, cross-layer design, security aspects and the on-time monitoring and discovery of topology changes are the main objectives and steps to be accomplished. Real-time Location monitoring and forest structure discovery are made possible by employing adaptive routing protocols, namely Dynamic Source Routing(DSR) and Ad-Hoc On-Demand Distance Vector(AODV). Also, the algorithms are capable of making advance adjustments to routing routes and resource allocations drawn from the mobility patterns forecasting which is vital. Load balancing methods, which distribute traffic in a network evenly, enable the system to run efficiently and avoid problems caused by overloading. Algorithms that are energy-efficient and make the best use of energy resources having their topology's optimized are introduced to the streamlined energy distribution systems. These algorithms are focused on the node's energy level. Through fault tolerance techniques which recognize and handle failures in nodes and partitioned networks, connectedness activity is a continuous one. Through QoS perception, the network topology might automatically be tailored to conform to the given quality criteria, thus not restricting the choice for differing applications. The proposed could have defenses against adversary nodes and their potential DDoS attacks which will enhance network protection. Yet another advantage of an architectural style that promotes cross-layer collaboration is enhanced topology management which is the goal around which the functions of various levels of the protocol stack revolve

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Semantic-Aware Pseudo-labeling for Unsupervised Meta-Learning

引用

IEEE Transactions on Pattern Analysis and Machine intelligence 2025年第7期47卷 5475-5488页

作者： Ouyang, Tianran Dong, Xingping Ye, Mang Du, Bo Shao, Ling Shen, Jianbing Wuhan University School of Computer Science National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan430072 China University of Chinese Academy of Sciences UCAS-Terminus AI Lab Beijing101408 China University of Macau State Key Laboratory of Internet of Things for Smart City Department of Computer and Information Science 314100 China

In unsupervised meta-learning, the clustering-based pseudo-labeling approach is an attractive framework, since it is model-agnostic, allowing it to synergize with supervised algorithms to learn from unlabeled data. However, the pseudo-labels suffer from clustering noise and semantic chaos problems, further impacting the effectiveness of meta-learning. In this paper, we analyze and optimize the pseudo-labeling process, including encoding and clustering, aiming to generate semantic-like pseudo-labels to narrow the gap between unsupervised and supervised meta-learning. First, during the encoding, we observe that the embedding space of existing methods lacks clustering-friendly properties, which is the primary reason for clustering noise. To address this issue, we minimize the inter-to-intra-class similarity ratio to generate clustering-friendly embedding features and validate our approach through comprehensive experiments. Then, during the clustering, we find that the semantic quality of pseudo-labels is not adequately controlled, resulting in semantic chaos of pseudo-labels. We propose a semantic-stability index to measure the semantic quality of pseudo-labels quantitatively. Based on this index, we propose the Semantic-aware Pseudo-label Reassignment mechanism to generate semantic-like pseudo-labels for all samples. Our approach is model-agnostic and can easily be integrated into existing supervised methods. To demonstrate its generalization ability, we integrate it into two representative algorithms: MAML and EP. The results on three main few-shot benchmarks clearly show that the proposed method achieves significant improvement compared to state-of-the-art models. Notably, our approach also outperforms the corresponding supervised method in three tasks. © 1979-2012 IEEE.

关键词： Latent semantic analysis

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：