检索结果-内蒙古大学图书馆

The Hierarchical Hybrid Fuzzy-Neural Network Based on Lasso Function and Its Application to Classification of Remote Sensing Images

引用

Chinese Journal of Geophysics 2013年第4期54卷 590-598页

作者： Xian-Chuan YU Sha DAI Dan HU Qi-Yu JIANG College of Information Science and Technology Beijing Normal University Beijing 100875 China Southeast Fujian Geologic Party Quanzhou 362021 China

In this paper, a new algorithm for the hierarchical hybrid fuzzy-neural network model is proposed. The Takagi-Sugeno model and triangular membership function are adopted in the fuzzy system, and the Lasso function of the coefficient contraction method is used to reduce the strong interaction among discrete input variables. In the end, an experimental test on surface features classification by using LANDSAT ETM+ remote sensing image data of Zhangping and Anxi in Fujian Province is conducted. Compared with other neural networks, the classification result with the proposed approach is the most accurate, which proves its feasibility and validity, and can be used as a new classification method for surface features on remote sensing images.

关键词： Takagi-Sugeno model Hierarchical hybrid fuzzy-neural network Lasso function training algorithm Remote sensing image classification

来源：评论

学校读者我要写书评

暂无评论

Designing Stepped Impedance Microstrip Low-Pass Filters Using Artificial Neural Network at 1.8 GHz

Designing Stepped Impedance Microstrip Low-Pass Filters Usin...

引用

International Conference on Communication Systems and Network Technologies

作者： Vivek Singh Kushwah Geetam S. Tomar Sarita Singh Bhadauria Dept. of Electronics Amity School of Engineering & Technology Gwalior India Machine Intelligence Research Labs Gwalior 474011 India Dept. of Electronics Madhav Institute of Technology& Science Gwalior India

ISBN: (纸本)9781467356039

In this paper a design technique for a Stepped impedance Microstrip Low-pass filters is presented by using the artificial neural network (ANN) modeling method. Required dimensions of the microstrip filter layout are used to get the input-output relationships in the ANN model. This paper presents the design and analysis of Stepped impedance Microstrip Low-pass filters at cut-off frequency 1.8 GHz which gives insertion loss (S_(21)) of -3 dB at cut-off point. Subsequently an artificial neural network model is developed to find out the Magnitude variation of scattering parameters (S-parameters) of Microstrip Low-pass filters at 1.8 GHz for different dimensions. After developing the ANN model of microstrip low-pass filter, it is computationally more efficient in the design and it has been displayed to be as accurate as an Electromagnetic simulator. The simulation is performed using the commercial software IE3D 14.1 and MATLAB programming language GUI Tools are used for ANN training.

关键词： Microstrip Low Pass Filters Lumped elements ANN model MATLAB IE3D EM Simulation S-parameters training algorithm

来源：评论

学校读者我要写书评

暂无评论

F-SVC: A SIMPLE AND FAST training algorithm SOFT MARGIN SUPPORT VECTOR CLASSIFICATION

F-SVC: A SIMPLE AND FAST TRAINING ALGORITHM SOFT MARGIN SUPP...

引用

IEEE Workshop on Machine Learning for Signal Processing

作者： Tohme, Mireille Lengelle, Regis FORENAP Frp 27 Rue 4Eme RSM F-68250 Rouffach France Univ technol Troyes Inst Charles Delaunay LM2S FRE CNRS 2848 F-10010 Troyes France

ISBN: (纸本)9781424423750

Support Vector Machines have obtained much success in machine learning. But their training require to solve a quadratic optimization problem so that training time increases dramatically with the increase of the training set size. Hence, standard SVM have difficulty in handling large scale problems. In this paper, we present a new fast training algorithm for soft margin Support Vector Classification. This algorithm searches for successive efficient feasible directions. A heuristic for searching the direction maximally correlated with the gradient is applied and the optimum step size of the optimization algorithm is analytically determined. Furthermore the solution, gradient and objective function are recursively obtained. In order to deal with large scale problems, the Gram matrix has not to be stored. Our iterative algorithm fully exploits quadratic functions properties. F-SVC is very simple, easy to implement and able to perform on large data sets.

关键词： training algorithm Support Vector Machines Classification

来源：评论

学校读者我要写书评

暂无评论

Optimal location of interline power flow controller for controlling multi transmission line： A new integrated technique

引用

Frontiers of Electrical and Electronic Engineering in China 2012年第4期7卷 447-458页

作者： B. KARTHIK I. ALAGARASAN S. CHANDRASEKAR Department of Electrical and Electronics Engineering Sona College ofTechnology Salem India Department of Electrical and Electronics Engineering Kavery Engineer-ing College Salem India

In this paper, an interline power flow controller （IPFC） is used for controlling multi transmission lines. However, the optimal placement of IPFC in the transmis-sion line is a major problem. Thus, we use a combination of tabu search （TS） algorithm and artificial neural network （ANN） in the proposed method to find out the best placement locations for IPFC in a given multi transmission line system. TS algorithm is an optimization algorithm and we use it in the proposed method to determine the optimum bus combination using line data. Then, using the optimum bus combination, the neural network is trained to find out the best placement locations for IPFC. Finally, IPFC is connected at the best locations indicated by the neural network. Furthermore, using Newton-Raphson load flow algorithm, the transmission line loss of the IPFC connected bus is analyzed. The proposed methodology is implemen- ted in MATLAB working platform and tested on the IEEE-14 bus system. The output is compared with the genetic algorithm （GA） and general load flow analysis. The results are validated with Levenberg-Marquardt back propagation and gradient descent with momentum network training algorithm.

关键词： IEEE-14 bus system interline power flowcontroller （IPFC） tabu search （TS） algorithm artificialneural network （ANN） training algorithm load flow

来源：评论

学校读者我要写书评

暂无评论

Evaluating the training Dynamics of a CMOS based Synapse

Evaluating the Training Dynamics of a CMOS based Synapse

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Ghani, Arfan McDaid, Liam J. Belatreche, Ammar Kelly, Peter Hall, Steve Dowrick, Tom Huang, Shou Marsland, John Smith, Andy Univ Ulster Intelligent Syst Res Ctr Magee Campus Derry BT48 7JL North Ireland Univ Liverpool Liverpool L693GJ Merseyside England

ISBN: (纸本)9781424496365

Recent work by the authors proposed compact low power synapses in hardware, based on the charge-coupling principle, that can be configured to yield a static or dynamic response. The focus of this work is to investigate the training dynamics of these synapses. Empirical models of the Post Synaptic Response (PSP), derived from hardware simulations, were developed and subsequently embedded into the MATLAB environment. A network of these synapses was then used to solve a benchmark problem using a well established training algorithm where the performance metric was convergence time, accuracy and weight range;the Spike Response Model (SRM) was used to implement point neurons. Results are presented and compared with standard synaptic responses.

关键词： CMOS based synapse CMOS logic circuits Equations Firing Hardware MATLAB environment Mathematical model Neurons SRM Silicon training charge-coupling principle hardware simulations learning (artificial intelligence) neural nets post synaptic response spike response model training algorithm

来源：评论

学校读者我要写书评

暂无评论

Reduced symmetric self-constructing fuzzy neural network beamforming detectors

引用

IET MICROWAVES ANTENNAS & PROPAGATION 2011年第6期5卷 676-684页

作者： Chang, Y-J Ho, C-L Natl Cent Univ Dept Commun Engn Tao Yuan 32054 Taiwan

Beamforming technology has been widely used in smart antenna systems that can increase the user's capacity and coverage in modern communication products. In this study, a powerful reduced symmetric self-constructing fuzzy neural network (RS-SCFNN) beamforming detector is proposed for multi-antenna-assisted systems. A novel training algorithm for the RS-SCFNN beamformer is proposed based on clustering of array input vectors and an adaptive minimum bit-error rate method. An inherent symmetric property of the array input signal space is exploited to make training procedure of RS-SCFNN more efficient than that of standard SCFNN. In addition, the required amount of fuzzy rules can be greatly reduced in the RS-SCFNN structure. Simulation results demonstrate that RS-SCFNN beamformer provides superior performance to the classical linear ones and the other non-linear ones (including symmetric radial basis function, SCFNN and S-SCFNN), especially when supporting a large amount of users in the rank-deficient multi-antenna-assisted system.

关键词： smart antenna systems rank-deficient multiantenna-assisted system nonlinear ones mobile communication error statistics RS-SCFNN beamformer Communications computing Neural computing techniques Digital signal processing Mobile radio systems Other topics in statistics adaptive antenna arrays Signal detection radial basis function networks symmetric radial basis function RS-SCFNN structure inherent symmetric property telecommunication computing RS-SCFNN beamforming detector beamforming technology reduced symmetric self-constructing fuzzy neural network beamforming detectors communication products Antenna arrays array signal processing array input signal space fuzzy rules training algorithm array input vectors multiantenna-assisted systems adaptive minimum bit-error rate method fuzzy neural nets training procedure

来源：评论

学校读者我要写书评

暂无评论

On Optimal Control Scheme Based on Feedforward and Inverse Models of Artificial Neural Network

On Optimal Control Scheme Based on Feedforward and Inverse M...

引用

International Conference on Computer Science and Education (CSE 2011)

作者： Qu, Dongcai Zhao, Guorong Cao, Dong Lu, Jianhua Lu, Binwen Dept Control Engn Naval Aeronaut Yantai 264001 Peoples R China Naval Fly Acad PT-125000 Huludao Peoples R China

ISBN: (纸本)9783642224171

The mostly optimum control questions were based on precise mathematical model of the controlled object at present, and only aimed at the few parameters implementation optimization control. Because it was very difficult or imposable that established precise mathematical model of actual controlled *** actual massive complex. unknown and indefinite nonlinear objects had many optimization parameters, it made conventional optimum control method no longer to be suitable to. In order to solve these problems, in the analysis optimum control principle foundation, the optimum control scheme based on the ANN's (Artificial Neural Network) feedforward / inverse model had designed. Basis on characteristics of nonlinear mapping and adaptive study of ANN's models and so on, the configuration of the ANN's feedforward and inverse models were designed. Used specialized training method of ANN models, and based on the EF(Exponential Forgetting) algorithm renewed covariance matrix, iterative training online to ANN's models were carried out, therefore obtained the ANN's models with the optimization configuration. Simulation results show that, designed optimum control structure scheme is reasonable, and ANN's controller may get good controlling effect.

关键词： Optimum control Neural Network(NN) training algorithm

来源：评论

学校读者我要写书评

暂无评论

Optimistic bias in the assessment of high dimensional classifiers with a limited dataset

Optimistic bias in the assessment of high dimensional classi...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Chen, Weijie Brown, David G. US FDA Silver Spring MD 20993 USA

ISBN: (纸本)9781424496365

It is commonly recognized that using the same dataset for training and testing the classifier introduces optimistic bias in estimating classifier performance. However, bias of the same kind may still exist even when independent datasets are used for training and testing a classifier. This problem is especially important in the setting of high dimensional feature space and limited data. Bioinformatics data is typically characterized by a tremendous amount of data per patient but from a limited number of patients. Often the entire data set is utilized in a "pre-training" stage during which the feature set is winnowed to a manageable number, and the parameters of the training algorithm are established. Subsequently the data is bifurcated into training and test sets;however, bias has already been introduced into the classifier development process. We investigate the significance of this bias by performing simulated gene expression experiments. We find that, for data with moderate intrinsic separability and modest sample size, any observed separation is due to selection bias introduced in the aforementioned pre-training process. For greater intrinsic separability, correct data hygiene, i.e., complete separation of development and validation data yields a positive result, but one far less impressive than that mistakenly obtained using incomplete data separation.

关键词： Breast cancer Classification algorithms Covariance matrix Measurement Signal to noise ratio Testing training bioinformatics bioinformatics data classifier development process gene expression genetics high dimensional classifier optimistic bias pattern classification training algorithm

来源：评论

学校读者我要写书评

暂无评论

On Optimal Control Scheme Based on Feedforward and Inverse Models of Artificial Neural Network

On Optimal Control Scheme Based on Feedforward and Inverse M...

引用

Advances in Information Technology and Education

作者： Dongcai Qu Guorong Zhao Dong Cao Jianhua Lu Binwen Lu Department of Control Engineering of Naval Aeronautical and Astronautical University The Naval Fly Academy

The mostly optimum control questions were based on precise mathematical model of the controlled object at present, and only aimed at the few parameters implementation optimization control. Because it was very difficult or imposable that established precise mathematical model of actual controlled *** actual massive complex, unknown and indefinite nonlinear objects had many optimization parameters, it made conventional optimum control method no longer to be suitable to. In order to solve these problems, in the analysis optimum control principle foundation, the optimum control scheme based on the ANN's(Artificial Neural Network) feedforward / inverse model had designed. Basis on characteristics of nonlinear mapping and adaptive study of ANN's models and so on, the configuration of the ANN's feedforward and inverse models were designed. Used specialized training method of ANN models, and based on the EF(Exponential Forgetting) algorithm renewed covariance matrix, iterative training online to ANN's models were carried out, therefore obtained the ANN's models with the optimization configuration. Simulation results show that, designed optimum control structure scheme is reasonable, and ANN's controller may get good controlling effect.

关键词： Optimum control Neural Network(NN) training algorithm

来源：评论

学校读者我要写书评

暂无评论

Weak Classifiers Selecting based on PSO in Ada Boost

Weak Classifiers Selecting based on PSO in Ada Boost

引用

2011 International Conference on Future Software Engineering and Multimedia Engineering(FSME 2011)

作者： Rui Li Jiurui Zhang Li Mao Dept. of Computer and Communication Lanzhou University of Technology

Weak classifiers selection plays an important role in face detection based on Ada Boost algorithmMore discriminative weak classifiers can not only reduce training time but also enhance classification accuracyIn this paper, a new weak classifiers selecting algorithm based on PSO is proposed to make improvement in weak classifier selectionFirst, we eliminate some less discriminative features in the feature formation stage to select weak classifiers, rather than select weak classifiers using all the featuresSecond, the weak classifiers are constructed by using PSO to select the best features and the best thresholds, and then combine these key weak classifiers into a more effective strong classifierExperimental results indicate that the method can effectively solve the problem of too much training time and gain higher training efficiency.

关键词： face detection training algorithm Particle Swarm Optimization Ada Boost

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：