检索结果-南通市图书馆

Comparative study of multilayer perceptron-stochastic gradient descent and gradient boosted trees for predicting daily suspended sediment load: The case study of the Mississippi River, U.S.

在线全文

学校读者我要写书评

暂无评论

International Journal of Sediment Research 2021年第4期36卷 512-523页

作者： Sadra Shadkani Akram Abbaspour Saeed Samadianfard Sajjad Hashemi Amirhosein Mosavi Shahab S.Band Department of Water Engineering Faculty of Agriculture University of Tabriz Environmental Quality Atmospheric Science and Climate Change Research Group Ton Duc Thang University Faculty of Environment and Labour Safety Ton Duc Thang University Institute of Research and Development Duy Tan University Future Technology Research Center College of Future National Yunlin University of Science and Technology

Monitoring sediment transport is essential for managing and maintaining *** of the sediment load in rivers is fundamental for the study of sediment movement,erosion,and flood *** the current study,three machine learning models-multi-layer perceptron(MLP),multi-layer perceptron-stochastic gradient descent(MLP-SGD),and gradient boosted tree(GBT)-were utilized to estimate the suspended sediment load(SSL) at the ***(SL) and Chester(CH) stations on the Mississippi River,*** evaluation criteria including the Correlation Coefficient(CC),Nash Sutcliffe Efficiency(NSE),Scatter Index(SI),and Willmott’s Index(WI) were utilized to evaluate the performance of the used models.A sensitivity analysis of the models to the input variables revealed that the current day discharge variable had the most effect on the SSL at both stations,but in the absence of current-day discharge data(Qt),a combination of input parameters including SSLt-3,SSLt-2,SSLt-1,Qt-3,Qt-2,Qt-1can be used to estimate the *** comparative outcomes indicated the high accuracy of MLP-SGD-5 model with a CC of 0.983,SI of 0.254,WI of 0.991,and NSE of 0.967 at station CH and the MLP-SGD-6 model with a CC of 0.933,SI of 0.576,WI of 0.961,and NSE of 0.867,respectively,at station *** results of MLP models were improved by SGD ***,the MLP-SGD method is recommended as the most accurate model for SSL estimation.

关键词： Suspended sediment transport Ensemble learning Multilayer perceptron stochastic gradient descent Machine learning

Differentially private SGD with random features

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Applied Mathematics(A Journal of Chinese Universities) 2024年第1期39卷 1-23页

作者： WANG Yi-guang GUO Zheng-chu Polytechnic Institute of Zhejiang University Zhejiang UniversityHangzhou 310015China School of Mathematical Sciences Zhejiang UniversityHangzhou 310058China

In the realm of large-scale machine learning,it is crucial to explore methods for reducing computational complexity and memory demands while maintaining generalization ***,since the collected data may contain some sensitive information,it is also of great significance to study privacy-preserving machine learning *** paper focuses on the performance of the differentially private stochastic gradient descent(SGD)algorithm based on random *** begin,the algorithm maps the original data into a lowdimensional space,thereby avoiding the traditional kernel method for large-scale data storage ***,the algorithm iteratively optimizes parameters using the stochastic gradient descent ***,the output perturbation mechanism is employed to introduce random noise,ensuring algorithmic *** prove that the proposed algorithm satisfies the differential privacy while achieving fast convergence rates under some mild conditions.

关键词： learning theory differential privacy stochastic gradient descent random features reproducing kernel Hilbert spaces

FL-EASGD:Federated Learning Privacy Security Method Based on Homomorphic Encryption

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第5期79卷 2361-2373页

作者： Hao Sun Xiubo Chen Kaiguo Yuan Information Security Center State Key Laboratory ofNetworking and Switching TechnologyBeijing University of Posts and TelecommunicationsBeijing100876China

Federated learning ensures data privacy and security by sharing models among multiple computing nodes instead of plaintext ***,there is still a potential risk of privacy leakage,for example,attackers can obtain the original data through model inference ***,safeguarding the privacy of model parameters becomes *** proposed solution involves incorporating homomorphic encryption algorithms into the federated learning ***,the existing federated learning privacy protection scheme based on homomorphic encryption will greatly reduce the efficiency and robustness when there are performance differences between parties or abnormal *** solve the above problems,this paper proposes a privacy protection scheme named Federated Learning-Elastic Averaging stochastic gradient descent(FL-EASGD)based on a fully homomorphic encryption ***,this paper introduces the homomorphic encryption algorithm into the FL-EASGD scheme to preventmodel plaintext leakage and realize privacy security in the process ofmodel ***,this paper designs a robust model aggregation algorithm by adding time variables and constraint coefficients,which ensures the accuracy of model prediction while solving performance differences such as computation speed and node anomalies such as downtime of each *** addition,the scheme in this paper preserves the independent exploration of the local model by the nodes of each party,making the model more applicable to the local data ***,experimental analysis shows that when there are abnormalities in the participants,the efficiency and accuracy of the whole protocol are not significantly affected.

关键词： Federated learning homomorphic encryption privacy security stochastic gradient descent

A Primal-Dual SGD Algorithm for Distributed Nonconvex Optimization

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2022年第5期9卷 812-833页

作者： Xinlei Yi Shengjun Zhang Tao Yang Tianyou Chai Karl Henrik Johansson Division of Decision and Control Systems School of Electrical Engineering and Computer ScienceKTH Royal Institute of Technologyand also affiliated with the Digital FuturesStockholm 10044Sweden Department of Electrical Engineering University of North TexasDentonTX 76203 USA IEEE State Key Laboratory of Synthetical Automation for Process Industries Northeastern UniversityShenyang 110819China

The distributed nonconvex optimization problem of minimizing a global cost function formed by a sum of n local cost functions by using local information exchange is *** problem is an important component of many machine learning techniques with data parallelism,such as deep learning and federated *** propose a distributed primal-dual stochastic gradient descent(SGD)algorithm,suitable for arbitrarily connected communication networks and any smooth(possibly nonconvex)cost *** show that the proposed algorithm achieves the linear speedup convergence rate O(1/(√nT))for general nonconvex cost functions and the linear speedup convergence rate O(1/(nT)) when the global cost function satisfies the Polyak-Lojasiewicz(P-L)condition,where T is the total number of *** also show that the output of the proposed algorithm with constant parameters linearly converges to a neighborhood of a global *** demonstrate through numerical experiments the efficiency of our algorithm in comparison with the baseline centralized SGD and recently proposed distributed SGD algorithms.

关键词： Distributed nonconvex optimization linear speedup Polyak-Lojasiewicz(P-L)condition primal-dual algorithm stochastic gradient descent

Routing with Cooperative Nodes Using Improved Learning Approaches

在线全文

学校读者我要写书评

暂无评论

Intelligent Automation & Soft Computing 2023年第3期35卷 2857-2874页

作者： R.Raja N.Satheesh J.Britto Dennis C.Raghavendra Department of CSIT CVR College of EngineeringTelanganaIndia Department of Computer Science and Engineering St.Martin’s Engineering CollegeSecunderabadTelanganaIndia Department of Computer Science and Engineering Dhanalakshmi Srinivasan UniversityTrichyTamilnaduIndia CSIT Department CVR College of EngineeringHyderabadIndia

In IoT,routing among the cooperative nodes plays an incredible role in fulﬁlling the network requirements and enhancing system *** eva-luation of optimal routing and related routing parameters over the deployed net-work environment is *** research concentrates on modelling a memory-based routing model with Stacked Long Short Term Memory(s-LSTM)and Bi-directional Long Short Term Memory(b-LSTM).It is used to hold the routing information and random routing to attain superior *** pro-posed model is trained based on the searching and detection mechanisms to com-pute the packet delivery ratio(PDR),end-to-end(E2E)delay,throughput,*** anticipated s-LSTM and b-LSTM model intends to ensure Quality of Service(QoS)even in changing network *** performance of the proposed b-LSTM and s-LSTM is measured by comparing the signiﬁcance of the model with various prevailing ***,the performance is measured with Mean Absolute Error(MAE)and Root Mean Square Error(RMSE)for mea-suring the error rate of the *** prediction of error rate is made with Learn-ing-based stochastic gradient descent(L-SGD).This gradual gradient descent intends to predict the maximal or minimal error through successive *** simulation is performed in a MATLAB 2020a environment,and the model performance is evaluated with diverse *** anticipated model intends to give superior performance in contrast to prevailing approaches.

关键词： Internet of Things(IoT) stacked long short term memory bi-directional long short term memory error rate stochastic gradient descent

A Sample-Wise Data Driven Control Solver for the stochastic Optimal Control Problem with Unknown Model Parameters

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Communications in Computational Physics 2023年第4期33卷 1132-1163页

作者： Richard Archibald Feng Bao Jiongmin Yong Computational Science and Mathematics Division Oak Ridge National LaboratoryOak RidgeTennesseeUSA Department of Mathematics Florida State UniversityTallahasseeFloridaUSA Department of Mathematics University of Central FloridaOrlandoFloridaUSA

In this work,an efficient sample-wise data driven control solver will be developed to solve the stochastic optimal control problem with unknown model parameters.A direct filter method will be applied as an online parameter estimation method that dynamically estimates the target model parameters upon receiving the data,and a sample-wise optimal control solver will be provided to efficiently search for the optimal ***,an effective overarching algorithm will be introduced to combine the parameter estimator and the optimal control *** experiments will be carried out to demonstrate the effectiveness and the efficiency of the sample-wise data driven control method.

关键词： stochastic optimal control parameter estimation optimal filter backward stochastic differential equations stochastic gradient descent

Chimp Optimization Algorithm Based Feature Selection with Machine Learning for Medical Data Classification

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computer Systems Science & Engineering 2023年第12期47卷 2791-2814页

作者： Firas Abedi Hayder M.A.Ghanimi Abeer D.Algarni Naglaa F.Soliman Walid El-Shafai Ali Hashim Abbas Zahraa H.Kareem Hussein Muhi Hariz Ahmed Alkhayyat Department of Mathematics College of EducationAl-Zahraa University for WomenKarbalaIraq Biomedical Engineering Department College of EngineeringUniversity ofWarith Al-AnbiyaaKarbalaIraq Department of Information Technology College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Computer Science Department Security Engineering LabPrince Sultan UniversityRiyadh11586Saudi Arabia Department of Electronics and Electrical Communications Engineering Faculty of Electronic EngineeringMenoufia UniversityMenouf32952Egypt College of Information Technology Imam Jaafar Al-Sadiq UniversityAl-Muthanna66002Iraq Department of Medical Instrumentation Techniques Engineering Al-Mustaqbal University CollegeHillah51001Iraq Computer Engineering Department Mazaya University CollegeDhi QarIraq College of Technical Engineering the Islamic UniversityNajafIraq

Datamining plays a crucial role in extractingmeaningful knowledge fromlarge-scale data repositories,such as data warehouses and *** rule mining,a fundamental process in data mining,involves discovering correlations,patterns,and causal structures within *** the healthcare domain,association rules offer valuable opportunities for building knowledge bases,enabling intelligent diagnoses,and extracting invaluable information *** paper presents a novel approach called the Machine Learning based Association Rule Mining and Classification for Healthcare Data Management System(MLARMC-HDMS).The MLARMC-HDMS technique integrates classification and association rule mining(ARM)***,the chimp optimization algorithm-based feature selection(COAFS)technique is employed within MLARMC-HDMS to select relevant *** by the foraging behavior of chimpanzees,the COA algorithm mimics their search strategy for ***,the classification process utilizes stochastic gradient descent with a multilayer perceptron(SGD-MLP)model,while the Apriori algorithm determines attribute *** propose a COA-based feature selection approach for medical data classification using machine learning *** approach involves selecting pertinent features from medical datasets through COA and training machine learning models using the reduced feature *** evaluate the performance of our approach on various medical datasets employing diverse machine learning *** results demonstrate that our proposed approach surpasses alternative feature selection methods,achieving higher accuracy and precision rates in medical data classification *** study showcases the effectiveness and efficiency of the COA-based feature selection approach in identifying relevant features,thereby enhancing the diagnosis and treatment of various *** provide further validation,we conduct detailed experiments on a benchmark

关键词： Association rule mining data classification healthcare data machine learning parameter tuning data mining feature selection MLARMC-HDMS COA stochastic gradient descent Apriori algorithm

Research on three-step accelerated gradient algorithm in deep learning

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Statistical Theory and Related Fields 2022年第1期6卷 40-57页

作者： Yongqiang Lian Yincai Tang Shirong Zhou KLATASDS-MOE School of StatisticsEast China Normal UniversityShanghaiPeople's Republic of China

gradient descent(GD)algorithm is the widely used optimisation method in training machine learning and deep learning *** this paper,based on GD,Polyak’s momentum(PM),and Nesterov accelerated gradient(NAG),we give the convergence of the algorithms from an ini-tial value to the optimal value of an objective function in simple quadratic *** on the convergence property of the quadratic function,two sister sequences of NAG’s iteration and par-allel tangent methods in neural networks,the three-step accelerated gradient(TAG)algorithm is proposed,which has three sequences other than two sister *** illustrate the perfor-mance of this algorithm,we compare the proposed algorithm with the three other algorithms in quadratic function,high-dimensional quadratic functions,and nonquadratic *** we consider to combine the TAG algorithm to the backpropagation algorithm and the stochastic gradient descent algorithm in deep *** conveniently facilitate the proposed algorithms,we rewite the R package‘neuralnet’and extend it to‘supneuralnet’.All kinds of deep learning algorithms in this paper are included in‘supneuralnet’***,we show our algorithms are superior to other algorithms in four case studies.

关键词： Accelerated algorithm backpropagation deep learning learning rate momentum stochastic gradient descent

Neural-network-based Power System State Estimation with Extended Observability

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 Modern Power Systems and Clean Energy 2021年第5期9卷 1043-1053页

作者： Guanyu Tian Yingzhong Gu Di Shi Jing Fu Zhe Yu Qun Zhou Department of Electrical and Computer Engineering University of Central FloridaOrlandoFL 32816USA GEIRI North America San JoseCA 95134USA State Grid Jiangsu Electric Power Company NanjingChina

This paper proposes a neural-network-based state estimation(NNSE)method that aims to achieve higher time efficiency,improved robustness against noise,and extended observability when compared with the conventional weighted least squares(WLS)state estimation *** consists of two parts,the linear state estimation neural network(LSE-net)and the unobservable state estimation neural network(USE-net).The LSE-net functions as an adaptive approximator of linear state estimation(LSE)equations to estimate the nominally observable *** inputs of LSE-net are the vectors of synchrophasors while the outputs are the estimated *** USE-net operates as the complementary estimator on the nominally unobservable *** inputs are the estimated observable states from LSE-net while the outputs are the estimation of nominally unobservable ***-net is trained off-line to approximate the veiled relationship between observable states and unobservable *** test cases are conducted to validate the performance of the proposed *** first case,which is based on the IEEE 118-bus system,shows the comprehensive performance of convergence,accuracy,and robustness of the proposed *** second case study adopts real-world synchrophasor measurements,and is based on the Jiangsu power grid,which is one of the largest provincial power systems in China.

关键词： State estimation linear state estimation stochastic gradient descent neural network wide area management system(WAMS).