Your Cart

Your cart is empty

Home  >  Volume 22 (2012)

On R^2 Contribution and Statistical Inference of the Change in the Hidden and Input Units Of The Statistical Neural Networks by Christopher Godwin Udomboso, 2Tolulope Olayemi James and 3Mba Obasi Odim Volume 22 (November, 2012), pp 335 – 340
Sale price: $5.00
Quantity:

Abstract

Determining the number of hidden units for obtaining optimal network performance has been a concern over the years despite empirical results showing that with higher neurons, the network error is reduced.  This has led to indiscrimate increase in the hidden neurons, thereby bringing about overfitting.  On the other hand, using too few hidden neurons leads to error bias, which can make neural network statistically unfit.  In this paper, we developed a model for R^2 for investigating changes in hidden and input units, as well as developed tests that can be used in determining the number of hidden and input units to obtain optimal performance.  The result of the analyses shows that there is effect on the network model when there is an increase in the number of hidden neurons, as well as the number of input units. 

Keywords: Hidden Unit, Input Unit, R^2 change, F test.

Close
Loading...