A Gaussian potential function network with hierarchically self-organizing learning

Lee, Sukhan; Kil, Rhee Man

doi:10.1016/0893-6080(91)90005-p

Cited by 305 publications

(81 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The algorithm is based on the idea that the number of hidden units should correspond to the complexity of the underlying function as re#ected in the observed data. Lee et al [10] developed hierarchically self-organizing learning (HSOL) in order to determine the optimal number of hidden units of their Gaussian function network. For the same purpose, Musavi et al [13] employed a method in which a large number of hidden nodes are merged whenever possible.…”

Section: Sequential Learning Using the Pg-rbf Networkmentioning

confidence: 99%

Time series analysis using normalized PG-RBF network with regression weights

et al. 2002

View full text Add to dashboard Cite

This paper proposes a framework for constructing and training a radial basis function (RBF) neural network. For this purpose, a sequential learning algorithm is presented to adapt the structure of the network, in which it is possible to create a new hidden unit and also to detect and remove inactive units. The structure of the Gaussian functions is modi"ed using a pseudoGaussian function (PG) in which two scaling parameters are introduced, which eliminates the symmetry restriction and provides the neurons in the hidden layer with greater #exibility with respect to function approximation. Other important characteristics of the proposed neural system are that the activation of the hidden neurons is normalized which, as described in the bibliography, provides a better performance than nonnormalization and instead of using a single parameter for the output weights, these are functions of the input variables which leads to a signi"cant reduction in the number of hidden units compared to the classical RBF network. Finally, we examine the result of applying the proposed algorithm to time series prediction.

show abstract

Section: Sequential Learning Using the Pg-rbf Networkmentioning

confidence: 99%

Time series analysis using normalized PG-RBF network with regression weights

et al. 2002

View full text Add to dashboard Cite

show abstract

“…Since recurrent networks incorporate feedback, they have powerful representation capability and can successfully overcome disadvantages of feed forward networks [8]. RBF neural networks have been used as a powerful tool in many engineering andscientific applications as they possess the following features: 1) They are universal approximators [1]; 2) They have a simple topological structure [2]; 3) They can implement fast learning algorithms because of locally tuned neurons [3]. In this study a recurrent RBF neural network is introduced where the RBF network"s ability is added to the advantages of recurrent networks.…”

Section: Introductionmentioning

confidence: 99%

The Proposal of Two New Recurrent Radial Basis Function Neural Networks

Shafiabady¹,

Isa²,

Vakilian³

2014

IJCA

View full text Add to dashboard Cite

show abstract

“…In addition, these approaches usually work offline, so they are not suitable for practical real-time applications where the online learning is required for the neural-network-based controller design. To remedy the aforementioned shortcomings, several growing RBF networks have been proposed in [8,11,12,18,21].…”

Section: Introductionmentioning

confidence: 99%

Variable Structure Neural Networks for Adaptive Robust Control Using Evolutionary Artificial Potential Fields

Mekki¹,

Chtourou²

2013

Journal of Electrical Engineering

View full text Add to dashboard Cite

A novel neural network architecture, is proposed and shown to be useful in approximating the unknown nonlinearities of dynamical systems. In the variable structure neural network, the number of basis functions can be either increased or decreased with time according to specified design strategies so that the network will not overfit or underfit the data set. Based on the Gaussian radial basis function (GRBF) variable neural network, an adaptive state feedback controller is presented. The location of the centers of the GRBFs is analyzed using a new method inspired from evolutionary artificial potential fields method combined with a pruning algorithm. Using this method we can guarantee a minimal number of neuron. It is in noted, that both the recruitment and the pruning is made by a single neuron. Consequently, the recruitment phase does not perturb the network and the pruning does not provoke an oscillation of the output response. The weights of neural network are adapted using a Lyapunov approach. Moreover, the stability of the system can be analyzed and guaranteed by introducing the supervisory controller and modified adaptation law with projection.

show abstract

A Gaussian potential function network with hierarchically self-organizing learning

Cited by 305 publications

References 12 publications

Time series analysis using normalized PG-RBF network with regression weights

Time series analysis using normalized PG-RBF network with regression weights

The Proposal of Two New Recurrent Radial Basis Function Neural Networks

Variable Structure Neural Networks for Adaptive Robust Control Using Evolutionary Artificial Potential Fields

Contact Info

Product

Resources

About