Friday 29th of March 2024
 

Oscillation Characteristics of the Multi-stage Learning for the Layered Neural Networks and Its Analysis


Isao Taguch and Yasuo Sugai

This paper proposes an efficient learning method for the layered neural networks based on the selection of training data and input characteristics of an output layer unit. The multi-stage learning method proposes by the authors for the function approximation problems of classifying learning data in a phased manner, focusing on their learnabilities prior to learning in the multi layered neural network, and demonstrates validity of the multi-stage learning method. Specifically, this paper verifies by computer experiments that both of larning accuracy and learning time are improved of the BP method as a learning rule of the multi-stage learning method. The authors also discuss the occurrence mechanisms of oscillatory phenomena in learning. Furthermore, the authors discuss the reasons that errors of some data remain large value even after learning, observing behaviors during learning.

Keywords: data selection, function approximation problem, multi-stage leaning, neural network, voluntary oscillation.

Download Full-Text


ABOUT THE AUTHORS

Isao Taguch
1--5--21, Anagawa, Inage-ku, Chiba-shi, 263-8588, Japan

Yasuo Sugai
1-33, Yayoi-cho, Inage-ku, Chiba-shi, 263-8522, Japan


IJCSI Published Papers Indexed By:

 

 

 

 
+++
About IJCSI

IJCSI is a refereed open access international journal for scientific papers dealing in all areas of computer science research...

Learn more »
Join Us
FAQs

Read the most frequently asked questions about IJCSI.

Frequently Asked Questions (FAQs) »
Get in touch

Phone: +230 911 5482
Email: info@ijcsi.org

More contact details »