In order to construct a nonlinear regression model we have to accurately (in some sense) initialize parameters of the model. In this work we performed comparison of several widely used methods and several newly develo...
详细信息
In order to construct a nonlinear regression model we have to accurately (in some sense) initialize parameters of the model. In this work we performed comparison of several widely used methods and several newly developed approached for initialization of parameters of a regression model, represented as a decomposition in a linear dictionary of some parametric functions (sigmoids). We proposed a general deterministic approach for initialization, providing repeatability of results, reduction of a learning time and in some cases increase of a regression model accuracy;we developed two new algorithms (based on a piecewise-linear approximation and based on local properties of approximable dependency) in the framework of the proposed approach;we developed randomized initialization algorithm (spherical initialization) for effective approximation of high-dimensional dependencies;we improved the classical initialization method SCAWI (by locating centers of sigmoids in sample points), providing a regression model accuracy improvement on specific classes of dependencies (smooth functions and discontinuous functions with a number of local peculiarities in an input domain) when using rprop algorithm for learning;we performed comparison of classical and newly proposed initialization methods and highlighted the most efficient ones.
BCI (called Brain-Computer Interface) is an interface that allows direct communication between human brain and an external device. It bases on EEG signal collection, processing and classification. In this paper a comp...
详细信息
ISBN:
(纸本)9780819491718
BCI (called Brain-Computer Interface) is an interface that allows direct communication between human brain and an external device. It bases on EEG signal collection, processing and classification. In this paper a complete BCI system is presented which classifies EEG signal using artificial neural networks. For this purpose we used a multi-layered perceptron architecture trained with the rprop algorithm. Furthermore a simple multi-threaded method for automatic network structure optimizing was shown. We presented the results of our system in the opening and closing eyes recognition task. We also showed how our system could be used for controlling devices basing on imaginary hand movements.
Blind equalization by neural network has two difficult problems, which is convergence rate and computational complexity. Resilient BP algorithm (rprop) combining compressed transfer function is proposed to Improve bli...
详细信息
ISBN:
(纸本)9781424421077
Blind equalization by neural network has two difficult problems, which is convergence rate and computational complexity. Resilient BP algorithm (rprop) combining compressed transfer function is proposed to Improve blind equalization by neural network. Compressed transfer function can make the Input signal avoid saturation zone and rprop algorithm can improve convergence rate effectively without adding additional calculation amount. The effectiveness of the algorithm is identified by simulation.
This paper demonstrates how the p-recursive piecewise polynomial (p-RPP) generators. and their derivatives are constructed. The feedforward computational time of a multilayer feedforward network can be reduced by usin...
详细信息
This paper demonstrates how the p-recursive piecewise polynomial (p-RPP) generators. and their derivatives are constructed. The feedforward computational time of a multilayer feedforward network can be reduced by using these functions as the activation functions. Three modifications of training algorithms are proposed. First, we use the modified error function so that the sigmoid prime factor for the updating rule of the output units is eliminated. Second, we normalize the input patterns in order to balance the dynamic range of the inputs. And third, we add a new penalty function to the hidden layer to get the anti-Hebbian rules in providing information when the activation functions have zero sigmoid prime factor. The three modifications are combined with two versions of rprop (Resilient propagation) algorithm. The proposed procedures achieved excellent results without the need for careful selection of the training parameters. Not only the algorithm but also the shape of the activation function has important influence on the training performance.
This paper is devoted to the field of Artificial Intelligence for drive control. In previous works, we presented possible advantages from using an Artificial Neural Network (ANN) for speed control in a DTC-SVM (Direct...
详细信息
ISBN:
(纸本)9781424408122
This paper is devoted to the field of Artificial Intelligence for drive control. In previous works, we presented possible advantages from using an Artificial Neural Network (ANN) for speed control in a DTC-SVM (Direct Torque Controlled-Space Vector Modulated) drive. Learning of the neural controller was set on-line. Starting from a random configuration of the speed controller, the network adapts its weights according to an error criterion. Although the use of such specialized controller allows potential adaptive and robust control skills, tuning of an ANN for online learning control is a long iterative procedure. Indeed, optimization of the neural controller induces determination of ten parameters acting critically on the control dynamics. However, using optimization algorithms, one can reduce efforts to reveal this set of parameters. Several optimization algorithms are based on description of biological evolutions. We call such algorithms Evolutionary algorithms (EA). Genetic algorithm (GA) is a EA inspired by genetic processes leading human race toward optimal individuals capable of controlling their environment. This paper presents GA for optimization of ANN-based speed controller for induction motor drive.
暂无评论