Siraj, Fadzilah and Partridge, Derek (2000) Systematic treatment of failures using multilayer perceptrons. In: Proceedings of the Thirteenth International Florida Artificial Intelligence Research Society Conference, 22-24 May 2000, Orlando, Florida USA.

PDF Restricted to Repository staff only Download (96kB)

Abstract

This paper discusses the empirical evaluation of improving generalization performance of neural networks by systematic treatment of training and test failures. As a result of systematic treatment of failures, multilayer perceptron (MLP) discriminants were developed as discrimination techniques. The experiments presented in this paper illustrate the application of discrimination techniques using MLP discriminants to neural networks trained to solve supervised learning task such as the Launch Interceptor Condition 1 problem. The MLP discriminants were constructed from the training and test patterns. The first discriminant is known as the hard-to-learn and easy-to-learn discriminant whilst the second one is known as hard-to-compute and easy-to-compute discriminant. Further treatments were also applied to hard-tolearn (or hard-to-compute) patterns prior to training (or testing). The experimental results reveal that directed splitting or using MLP discriminant is an important strategy in improving generalization of the networks.

Item Type:	Conference or Workshop Item (Paper)
Subjects:	Q Science > QA Mathematics > QA76 Computer software
Divisions:	College of Arts and Sciences
Depositing User:	Prof Madya Fadzilah Siraj
Date Deposited:	12 Nov 2010 06:58
Last Modified:	12 Nov 2010 06:58
URI:	https://repo.uum.edu.my/id/eprint/1540

Actions (login required)

View Item