检索结果-南通市图书馆

De-biased knowledge distillation framework based on knowledge infusion and label de-biasing techniques

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of electronic Science and Technology 2024年第3期22卷 57-68页

作者： Yan Li Tai-Kang Tian Meng-Yu Zhuang Yu-Ting Sun school of Economics and Management University of Electronic Science and Technology of ChinaChengdu611731China school of Economics and Management Beijing University of Posts and TelecommunicationBeijing100876China school of Electrical {1. and computer Science The University of QueenslandBrisbane4072Australia

Knowledge distillation,as a pivotal technique in the field of model compression,has been widely applied across various ***,the problem of student model performance being limited due to inherent biases in the teacher model during the distillation process still *** address the inherent biases in knowledge distillation,we propose a de-biased knowledge distillation framework tailored for binary classification *** the pre-trained teacher model,biases in the soft labels are mitigated through knowledge infusion and label de-biasing *** on this,a de-biased distillation loss is introduced,allowing the de-biased labels to replace the soft labels as the fitting target for the student *** approach enables the student model to learn from the corrected model information,achieving high-performance deployment on lightweight student *** conducted on multiple real-world datasets demonstrate that deep learning models compressed under the de-biased knowledge distillation framework significantly outperform traditional response-based and feature-based knowledge distillation models across various evaluation metrics,highlighting the effectiveness and superiority of the de-biased knowledge distillation framework in model compression.

关键词： De-biasing Deep learning Knowledge distillation Model compression

Optimized Binary Neural Networks for Road Anomaly Detection:A TinyML Approach on Edge Devices

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

computers, Materials & Continua 2024年第7期80卷 527-546页

作者： Amna Khatoon Weixing Wang Asad Ullah Limin Li Mengfei Wang school of Information {1. Chang’an UniversityXi’an710064China school of Information {1. Xi’an Eurasia UniversityXi’an710065China school of Electrical and {1. engineering Wenzhou UniversityWenzhou325035China

Integrating Tiny Machine Learning(TinyML)with edge computing in remotely sensed images enhances the capabilities of road anomaly detection on a broader *** devices efficiently implement a Binary Neural Network(BNN)for road feature extraction,utilizing quantization and compression through a pruning *** modifications resulted in a 28-fold decrease in memory usage and a 25%enhancement in inference speed while only experiencing a 2.5%decrease in *** showcases its superiority over conventional detection algorithms in different road image *** constrained by computer resources and training datasets,our results indicate opportunities for future research,demonstrating that quantization and focused optimization can significantly improve machine learning models’accuracy and operational *** Cortex-M0 gives practical feasibility and substantial benefits while deploying our optimized BNN model on this low-power device:Advanced machine learning in edge *** analysis work delves into the educational significance of TinyML and its essential function in analyzing road networks using remote sensing,suggesting ways to improve smart city frameworks in road network assessment,traffic management,and autonomous vehicle navigation systems by emphasizing the importance of new technologies for maintaining and safeguarding road networks.

关键词： Edge computing remote sensing TinyML optimization BNNs road anomaly detection quantization model compression

Investigation of wake-induced vibration interference between two staggered wave-cone cylinders at subcritical Reynolds number

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Hydrodynamics 2024年第2期36卷 378-393页

作者： Lin Zou Di-wei Liu Jian Liu Hong-cheng Zuo Yin-huan Zheng Hao Xia school of Mechanical and {1. engineering Wuhan University of TechnologyWuhan430070China Department of Aeronautical and Automotive engineering Loughborough UniversityLoughboroughUK

The wake induced vibration(WIV)of a one-and two-degree-of-freedom(1.OF,2DOF)downstream wave-cone cylinder(WCC)behind a stationary equal-size upstream wave-cone cylinder in the staggered arrangement is numerically investigated at subcritical Reynolds number of 3900 by using shear stressed transfer(SST)k-ωturbulence *** streamwise pitch ratios(P/Dm)vary from 4 to 6 with a fixed incident angleα=8°.Experimental measurements were also performed for the validation of the present numerical *** is found that the largest vibration amplitude in crossflow direction occurred at P/Dm=4,Ur=8 with small difference of streamwise vibration at P/Dm=4,*** from single wavy-cone cylinder(SWCC),the downstream flexible one of a pair staggered WCCs got larger vibration amplitude during phase switching stage instead of in-phase *** upstream wake will suppress the triple frequency of main frequency in the power spectra density(PSD)functions of Cl but stimulate the double one of *** intriguing vibration mechanism happened in all 2DOF cases where the trajectory of downstream WCC is a significant ellipse rather than a figure of *** transformation of phase switching and the variation of the main frequency of drag coefficient(Cd)can be explained by the vortex-shedding modes of downstream WCC.

关键词： Two wave-cone cylinders staggered arrangement wake induced vibration phase switching vortex shedding

Lightweight Malicious Code Classification Method Based on Improved Squeeze Net

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

computers, Materials & Continua 2024年第1期78卷 551-567页

作者： Li Li Youran Kong Qing Zhang school of {1. and Control engineering Northeast Forestry UniversityHarbin150040China

With the growth of the Internet,more and more business is being done online,for example,online offices,online education and so *** this makes people’s lives more convenient,it also increases the risk of the network being attacked by malicious ***,it is important to identify malicious codes on computer systems ***,most of the existing malicious code detection methods have two problems:(1.The ability of the model to extract features is weak,resulting in poor model performance.(2)The large scale of model data leads to difficulties deploying on devices with limited ***,this paper proposes a lightweight malicious code identification model Lightweight Malicious Code Classification Method Based on Improved SqueezeNet(LCMISNet).In this paper,the MFire lightweight feature extraction module is constructed by proposing a feature slicing module and a multi-size depthwise separable convolution *** feature slicing module reduces the number of parameters by grouping *** multi-size depthwise separable convolution module reduces the number of parameters and enhances the feature extraction capability by replacing the standard convolution with depthwise separable convolution with different convolution kernel *** addition,this paper also proposes a feature splicing module to connect the MFire lightweight feature extraction module based on the feature reuse and constructs the lightweight model *** malicious code recognition accuracy of LCMISNet on the BIG 201. dataset and the Malimg dataset reaches 98.90% and 99.58%,*** proves that LCMISNet has a powerful malicious code recognition *** addition,compared with other network models,LCMISNet has better performance,and a lower number of parameters and computations.

关键词： Lightweight neural network malicious code classification feature slicing feature splicing multi-size depthwise separable convolution

Modifying Lyapunov exponent of chaotic map by self-cascading

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science China(Technological Sciences) 2024年第7期67卷 2203-2214页

作者： YI ChenLong LI ChunBiao LI YongXin XIA Ming HUA ZhongYun school of {1. and Information engineering Nanjing University of Information Science&TechnologyNanjing 210044China school of Artificial Intelligence Nanjing University of Information Science&TechnologyNanjing 210044China school of {1. Science and Technology Harbin Institute of Technology Shenzhen Graduate SchoolShenzhen 518055China

The self-cascade(SC) method is an effective technique for chaos enhancement and complexity increasing in chaos ***, the controllable self-cascade(CSC) method allows for more accurate control of Lyapunov exponents of the discrete map. In this work, the SC and CSC systems of the original map are derived, which enhance the chaotic performance while preserving the fundamental dynamical characteristics of the original map. Higher Lyapunov exponent of chaotic sequences corresponding to higher frequency are obtained in SC and CSC systems. Meanwhile, the Lyapunov exponent could be linearly controlled with greater flexibility in the CSC system. The verification of the numerical simulation and theoretical analysis is carried out based on the platform of CH32.

关键词： self-cascade controllable self-cascade chaos enhancement chaos control

A survey on semantic communications:Technologies,solutions,applications and challenges

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Digital Communications and Networks 2024年第3期10卷 528-545页

作者： Yating Liu Xiaojie Wang Zhaolong Ning MengChu Zhou Lei Guo Behrouz Jedari school of Communication and Information {1. Chongqing University of Posts and TelecommunicationsChongqing 400065China Baotou Vocational&Technical College Baotou 014035China school of Information and {1. engineering Zhejiang Gongshang UniversityHangzhou 310018China Department of Electrical and computer {1. New Jersey Institute of TechnologyNewarkNJ 07102USA Nokia Finland

Semantic Communication(SC)has emerged as a novel communication paradigm that provides a receiver with meaningful information extracted from the source to maximize information transmission throughput in wireless networks,beyond the theoretical capacity *** the extensive research on SC,there is a lack of comprehensive survey on technologies,solutions,applications,and challenges for *** this article,the development of SC is first reviewed and its characteristics,architecture,and advantages are ***,key technologies such as semantic extraction,semantic encoding,and semantic segmentation are discussed and their corresponding solutions in terms of efficiency,robustness,adaptability,and reliability are *** of SC to UAV communication,remote image sensing and fusion,intelligent transportation,and healthcare are also presented and their strategies are ***,some challenges and future research directions are presented to provide guidance for further research of SC.

关键词： Semantic communication Semantic coding Semantic extraction Semantic communication framework Semantic communication applications

Effect of granular shape on radial segregation in a two-dimensional drum

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Physics B 2024年第4期33卷 509-514页

作者：徐悦李然迟志鹏修文正孙其诚杨晖 school of Optical-Electrical and {1. engineering University of Shanghai for Science and TechnologyShanghai 200093China school of Electrical and {1. engineering Shandong University of TechnologyZibo 255049China State Key Laboratory of Hydroscience and engineering Tsinghua UniversityBeijing 100084China College of Medical Instruments Shanghai University of Medicine and Health SciencesShanghai 201318China

Granular segregation is widely observed in nature and *** research has focused on segregation caused by differences in the size and density of spherical ***,due to the fact that grains typically have different shapes,the focus is shifting towards shape *** this study,experiments are conducted by mixing cubic and spherical *** results indicate that spherical grains gather at the center and cubic grains are distributed around them,and the degree of segregation is *** experiments,a structured analysis of local regions is conducted to explain the inability to form stable segregation patterns with obviously different geometric ***,through simulations,the reasons for the central and peripheral distributions are explained by comparing velocities and the number of collisions of the grains in the flow layer.

关键词： granular materials cubic grains shape segregation segregation mechanism

Distributed field mapping for mobile sensor teams using a derivative‐free optimisation algorithm

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

IET Cyber-Systems and Robotics 2024年第2期6卷 20-34页

作者： Tony X.Lin Jia Guo Said Al-Abri Fumin Zhang Fumin Zhang31.0} of Electrical&{1. engineering Georgia Institute of TechnologyAtlantaGeorgiaUSA College of engineering Sultan Qaboos UniversityMuscatOman Department of electronic&{1. engineering Hong Kong University of Science and TechnologyHong KongChina

The authors propose a distributed field mapping algorithm that drives a team of robots to explore and learn an unknown scalar field using a Gaussian Process(GP).The authors’strategy arises by balancing exploration objectives between areas of high error and high *** computing high error regions is impossible since the scalar field is unknown,a bio-inspired approach known as Speeding-Up and Slowing-Down is leveraged to track the gradient of the GP *** approach achieves global field-learning convergence and is shown to be resistant to poor hyperparameter tuning of the *** approach is validated in simulations and experiments using 2D wheeled robots and 2D flying mini-ature autonomous blimps.

关键词： environment sensing multi-robot systems

Robust Regularization Design of Graph Neural Networks Against Adversarial Attacks Based on Lyapunov Theory

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of electronics 2024年第3期33卷 732-741页

作者： Wenjie YAN Ziqi LI Yongjun QI school of Artificial Intelligence Hebei University of Technology school of {1. Science and engineering of North China Institute of Aerospace engineering

The robustness of graph neural networks（GNNs） is a critical research topic in deep *** researchers have designed regularization methods to enhance the robustness of neural networks,but there is a lack of theoretical analysis on the principle of *** order to tackle the weakness of current robustness designing methods,this paper gives new insights into how to guarantee the robustness of GNNs.A novel regularization strategy named Lya-Reg is designed to guarantee the robustness of GNNs by Lyapunov *** results give new insights into how regularization can mitigate the various adversarial effects on different graph *** experiments on various public datasets demonstrate that the proposed regularization method is more robust than the state-of-theart methods such as L1.norm,L2-norm,L2-norm,Pro-GNN,PA-GNN and GARNET against various types of graph adversarial attacks.

关键词： Training Deep learning Garnets Robustness Graph neural networks