The Joint Effect of factors on Generalization Performance of Neural Network Learning Procedure


The KIPS Transactions:PartB , Vol. 12, No. 3, pp. 343-348, Jun. 2005
10.3745/KIPSTB.2005.12.3.343,   PDF Download:

Abstract

The goal of this paper is to study the joint effect of factors of neural network learning procedure. There are many factors, which may affect the generalization ability and learning speed of neural networks, such as the initial values of weights, the learning rates, and the regularization coefficients. We will apply a constructive training algorithm for neural network, then patterns are trained incrementally by considering them one by one. First, we will investigate the effect of these factors on generalization performance and learning speed. Based on these factors´ effect, we will propose a joint method that simultaneously considers these three factors, and dynamically tune the learning rate and regularization coefficient. Then we will present the results of some experimental comparison among these kinds of methods in several simulated nonlinear data. Finally, we will draw conclusions and make plan for future work.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
Y. C. Yoon, "The Joint Effect of factors on Generalization Performance of Neural Network Learning Procedure," The KIPS Transactions:PartB , vol. 12, no. 3, pp. 343-348, 2005. DOI: 10.3745/KIPSTB.2005.12.3.343.

[ACM Style]
Yeo Chang Yoon. 2005. The Joint Effect of factors on Generalization Performance of Neural Network Learning Procedure. The KIPS Transactions:PartB , 12, 3, (2005), 343-348. DOI: 10.3745/KIPSTB.2005.12.3.343.