检索结果-内蒙古大学图书馆

IEEE National Aerospace and Electronics Conference (NAECON)

作者： Hasan, Raqibul Taha, Tarek M. Univ Dayton Dept Elect & Comp Engn Dayton OH 45469 USA

ISBN: (纸本)9781467375658

Several big data applications are particularly focused on classification and clustering tasks. Robustness of such system depends on how well it can extract important features from the raw data. For big data processing we are interested for a generic feature extraction mechanism for different applications. autoencoder is a popular unsupervised training algorithm for dimensionality reduction and feature extraction. In this work we have examined memiistor crossbar based implementation of autoencoder which will consume very low power. We have designed on-chip training circuitry for the unsupervised training scheme.

关键词： Low power architecture memristor crossbars autoencoder training algorithm feature extraction

来源：评论

学校读者我要写书评

暂无评论

Efficient Serial and Parallel SVM training using Coordinate Descent

Efficient Serial and Parallel SVM Training using Coordinate ...

引用

IEEE Symposium on Computational Intelligence for Engineering Solutions (CIES)

作者： Liossis, Emmanuel Natl Tech Univ Athens Sch Elect & Comp Engn Intelligent Syst Lab Athens Greece

ISBN: (纸本)9781467358514

Eliminating the bias term of the Support Vector Machine (SVM) classifier permits substancial simplification to training algorithms. Using this elimination, the optimization invloved in training can be decomposed to update as low as one coordinate at a time. This paper explores two directions of improvements which stem from this simplification. The first one is about the options available for choosing the coordinate to optimize during each optimization iteration. The second one is about the parallelization schemes which the simplified optimization facilitates.

关键词： SVM training algorithm parallel

来源：评论

学校读者我要写书评

暂无评论

Neural Network training based on quasi-Newton Method using Nesterov's Accelerated Gradient

Neural Network Training based on quasi-Newton Method using N...

引用

IEEE Region 10 Conference (TENCON)

作者： Ninomiya, Hiroshi Shonan Inst Technol Dept Informat Sci 1-1-25 Tsujido Nishikaigan Fujisawa Kanagawa 2518511 Japan

ISBN: (纸本)9781509025978

This paper describes a novel quasi-Newton (QN) based accelerated technique for training of neural networks. Recently, Nesterov's accelerated gradient method has been utilized for training the neural networks. In this paper the acceleration of the QN training algorithm is realized using Momentum term in Nesterov's method. It is shown that the proposed algorithm has a similar convergence property with the conventional QN method. Neural network training for a simple function approximation and a microwave circuit modeling is presented to demonstrate the proposed algorithm. The proposed algorithm drastically improves the convergence speed of the QN algorithm.

关键词： neural networks training algorithm Nesterov's accelerated gradient method quasi-Newton method

来源：评论

学校读者我要写书评

暂无评论

Architecture optimization, training convergence and network estimation robustness of a fully connected recurrent neural network

Architecture optimization, training convergence and network ...

引用

作者： Wang, Xiaoyu Clemson University

学位级别：Ph.D.

Recurrent neural networks (RNN) have been rapidly developed in recent years. Applications of RNN can be found in system identification, optimization, image processing, pattern reorganization, classification, clustering, memory association, etc. In this study, an optimized RNN is proposed to model nonlinear dynamical systems. A fully connected RNN is developed first which is modified from a fully forward connected neural network (FFCNN) by accommodating recurrent connections among its hidden neurons. In addition, a destructive structure optimization algorithm is applied and the extended Kalman filter (EKF) is adopted as a network’s training algorithm. These two algorithms can seamlessly work together to generate the optimized RNN. The enhancement of the modeling performance of the optimized network comes from three parts: (1) its prototype - the FFCNN has advantages over multilayer perceptron network (MLP), the most widely used network, in terms of modeling accuracy and generalization ability; (2) the recurrency in RNN network make it more capable of modeling non-linear dynamical systems; and (3) the structure optimization algorithm further improves RNN’s modeling performance in generalization ability and robustness. Performance studies of the proposed network are highlighted in training convergence and robustness. For the training convergence study, the Lyapunov method is used to adapt some training parameters to guarantee the training convergence, while the maximum likelihood method is used to estimate some other parameters to accelerate the training process. In addition, robustness analysis is conducted to develop a robustness measure considering uncertainties propagation through RNN via unscented transform. Two case studies, the modeling of a benchmark non-linear dynamical system and a tool wear progression in hard turning, are carried out to testify the development in this dissertation. The work detailed in this dissertation focuses on the creation of: (1) a new

关键词： Network Optimization Network Robustness Recurrent Neural Network training algorithm training Convergence

来源：评论

学校读者我要写书评

暂无评论

Using Particle Swarm Optimization in training Neural Network for Indoor Field Strength Prediction

Using Particle Swarm Optimization in Training Neural Network...

引用

51st International Symposium ELMAR

作者： Vilovic, Ivan Burum, Niksa Milic, Dorde Univ Dubrovnik Dubrovnik Croatia

ISBN: (纸本)9789537044107

This paper presents a comparison of results obtained from neural network training by backpropagation and particle swarm optimization (PSO) algorithms. The neural network model has been developed for field strength prediction in indoor environments. It has been already shown for neural networks as powerful tool in RF propagation prediction. It is very important to choose proper algorithm for training a neural network, so we compared BP training algorithms: gradient descent method and Levenberg- Marquardt algorithm with PSO algorithm. PSO algorithm has been shown as powerful method for global optimization in several applications. A floor of university building in Dubrovnik has been used as case for simulation and measurement of signal strength. The results show that the neural network weights converge faster with PSO than with standard BP algorithms.

关键词： Neural network Multilayer Perceptron training algorithm Backpropagation algorithm Gradient descent algorithm Levenberg-Marquardt algorithm Particle swarm optimization

来源：评论

学校读者我要写书评

暂无评论

A Novel Quasi-Newton-Based training Using Nesterov's Accelerated Gradient for Neural Networks 25th

A Novel Quasi-Newton-Based Training Using Nesterov's Acceler...

引用

25th International Conference on Artificial Neural Networks (ICANN)

作者： Ninomiya, Hiroshi Shonan Inst Technol Dept Informat Sci Fujisawa Kanagawa Japan

ISBN: (纸本)9783319447810;9783319447803

Neural networks have been recognized as a useful tool for the function approximation problems with high-nonlinearity [1]. training is the most important step in developing a neural network model. Gradient based algorithms are popularly used for the training. The popular methods are first-order methods so called Steepest Gradient (SG), Classical Momentum (CM), Nesterov’s Accelerated Gradient (NAG) methods and a more dedicated method such as RPROP. On the other hand, quasi-Newton (QN) method which is one of the most efficient optimization techniques with super-linear convergence is also widely utilized for more complicated problems. However, when applied to highly nonlinear function modeling, the above methods even QN still converges too slowly and optimization error cannot be effectively reduced within finite time. This paper describes a novel acceleration technique of QN using Nesterov’s accelerated gradient [2]. The proposed algorithm is referred to as Nesterov’s accelerated quasi-Newton (NAQ) method. The update equation of the approximated matrix of Hessian is mathematically derived from the quadratic approximation of error function. The Hessian is approximated using both of the normal gradient and the Nesterov’s accelerated one in NAQ. NAQ is a technique for accelerating QN that accumulates an update vector in directions of persistent reduction in the objective across iterations. As a result, the iteration during the NAQ training can be shortened without loss of the strong ability to search a global minimum in QN. The proposed algorithm is demonstrated through the computer simulations for the benchmark problems with high-nonlinearity compared with SG, CM, NAG, RPROP and QN. From the simulation results it is shown that NAQ is faster than the conventional methods without compromising quality of training solutions. © Springer International Publishing Switzerland 2016.

关键词： Neural networks training algorithm Nesterov's accelerated gradient method Quasi-Newton method

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

Robust training of Microwave Neural Network Models Using Combined Global/Local Optimization Techniques

Robust Training of Microwave Neural Network Models Using Com...

引用

2008 IEEE MTT-S International Microwave Symposium Digest

作者： Ninomiya, Hiroshi Wan, Shan Kabir, Humayun Zhang, Xin Zhang, Q. J. Shonan Inst Technol Dept Informat Sci Fujisawa Kanagawa 2518511 Japan Carleton Univ Dept Elect Ottawa ON K1S 5B6 Canada

ISBN: (纸本)9781424417803

We present a new technique for training microwave neural network models. The proposed technique combines quasi-Newton algorithm with a recent global optimization algorithm called Particle Swarm Optimization (PSO). The quasi-Newton process for searching optimal solutions is incorporated into PSO to speed up local search, while the PSO performs global search avoid being trapped in local minima of training. The overall algorithm iterates between quasi-Newton and PSO. Neural network training for waveguide and microstrip examples are presented, demonstrating that the proposed algorithm achieves more accurate models than the conventional gradient based technique and the conventional PSO.

关键词： Global optimization neural networks particle swarm optimization training algorithm

来源：评论

学校读者我要写书评

暂无评论

A training method for SpikeProp without redundant spikes -Removing unnecessary sub-connections during training- 12

A training method for SpikeProp without redundant spikes -Re...

引用

12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)

作者： Nakayama, Takutoshi Matsumoto, Takashi Takase, Haruhiko Kawanaka, Hiroharu Tsuruoka, Shinji Mie Univ Grad Sch Engn Tsu Mie 5148507 Japan Mie Univ Tsu Mie 5148507 Japan

ISBN: (纸本)9781509040933

SpikeProp, which is proposed by Bohte and extended by Booij, is a type of multi-layer networks of spiking neurons. Our research group has proposed a training algorithm for SpikeProp without redundant output spikes. However, its performance is sensitive to the initial network structure, such as the number of hidden units, the number of sub-connections, and their delays. In this article, we discuss the problem, especially the number of sub-connections. And we propose a method to remove unnecessary sub-connections during training. By simple experiments, we show that the method brings less dependency of training performance on the number of initial sub-connections.

关键词： Spiking Neural Network SpikeProp training algorithm Time Series Signal Processing

来源：评论

学校读者我要写书评

暂无评论

A new approach for spatio-temporal interface treatment in fluid-solid interaction using artificial neural networks employing coupled partitioned fluid-solid solvers

引用

JOURNAL OF FLUIDS AND STRUCTURES 2024年 131卷

作者： Mazhar, Farrukh Javed, Ali Natl Univ Sci & Technol Coll Aeronaut Engn Dept Aerosp Engn Islamabad 44000 Pakistan

Partitioned fluid-solid interaction (FSI) problems involving non-conforming grids pose formidable challenge in interface treatment, especially for information exchange, interface tracking, and field variable interpolation between solvers in both space and time. These demand special considerations for accurate and efficient simulations. This paper presents an application of artificial neural networks (ANN) for the interface treatment in a coupled FSI problem employing partitioned solvers. A shallow time-series ANN (nonlinear auto-regressive model with exogenous inputs, NARX) scheme is proposed to handle the exchange of Neumann/Dirichlet information at the coupling interface. This scheme involves two interface treatment models that were developed and analysed. The proposed models interpolate and transfer loads from the fluid to the solid domains, and conversely, displacements from the solid to the fluid domains between non-collocated grids. To validate this approach, we tested it on a 3D FSI problem, which involved damped oscillations of a flexible flap submerged in a fluid cavity. Adequately trained NARX interface models demonstrate reliable input-output mapping and accurate prediction of transient behaviour at the interface. Additionally, we explored the concept of reduced-order modelling (ROM) in the time domain. This allowed us to reduce the model's complexity by half. Different training algorithms were evaluated to enhance the efficiency and performance of the proposed scheme. The study demonstrates that NARX networks trained with Bayesian Regularization (BR) and Levenberg-Marquardt (LM) algorithms exhibit the best accuracy, while the scaled conjugate gradient (SCG)-based training method provides better computational efficiency with acceptable accuracy. Overall, the NARX interface models provide precise performance and offer a viable potential for applications in FSI problems requiring accurate and faster computations.

关键词： Fluid-solid interaction (FSI) Artificial neural network (ANN) Interface treatment training algorithm NARX neural network

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：