Symbolic function network

Eskander, George S.; Atiya, Amir F.

doi:10.1016/j.neunet.2009.02.003

Cited by 5 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The training data set has taken an incremental way to perform new data set. The algorithm is used the following steps for getting a forward network [13]; 1-Start from a blank network. Initialize all weights to zero.…”

Section: Proposed Methodsmentioning

confidence: 99%

“…After many attempts, it was found that the following parameters yield around 100% of training accuracy. The error measure used to assess the network performance is Mean Square Error (MSE) as follows [13]; where y (j), d (j) are the network output and the desired output j respectively, and J is the size of the data set. Most of error measures and the number of weights in the resulted network, which is given to compare the network complexities.…”

Section: Parameters Settingmentioning

confidence: 99%

See 1 more Smart Citation

Feed Forward Neural Network for Sine Function With Symmetric Table Addition Method Using Labview and Matlab Code

Ali¹

2014

IJCSA

View full text Add to dashboard Cite

This work is proposed the feed forward neural network with symmetric table addition method to design the neuron synapses algorithm of the sine function approximations, and according to the Taylor series expansion. Matlab code and LabVIEW are used to build and create the neural network, which has been designed and trained database set to improve its performance, and gets the best a global convergence with small value of MSE errors and 97.22% accuracy.

show abstract

Section: Proposed Methodsmentioning

confidence: 99%

Section: Parameters Settingmentioning

confidence: 99%

Feed Forward Neural Network for Sine Function With Symmetric Table Addition Method Using Labview and Matlab Code

Ali¹

2014

IJCSA

View full text Add to dashboard Cite

show abstract

“…That is a two layer network that consists of eight elementary functions with just fourteen tuned weights. This network representation is relatively simple compared to the simplest multilayer perceptron that could be designed to model this problem that contains more than sixty weights on average [1] Figure 3.The constructed symbolic function network.…”

Section: Synthetic Examplementioning

confidence: 99%

“…We propose an empirical approach that uses the collected RTT time series to construct a neural type predictor using a new neural network model called symbolic function network (SFN) [1]. This model is proved to have great approximation power and good ability to filter the system noise.…”

Section: Introductionmentioning

confidence: 99%