New genetic approach to universal rule generation from trained neural networks

Minoru Fukumi, Yasue Mitsukura, Norio Akamatsu

Research output: Contribution to conferencePaper

Abstract

In this paper a new rule generation method from neural networks is presented. A neural network (NN) is formed using a genetic algorithm (GA) with virus infection and deterministic mutation to represent regularities in training data. This method utilizes a modular structure in GA. Each module learns a different neural network architecture, such as sigmoid and a higher order neural networks. Those chromosome information is communicated to the other modules by the virus infection. The higher order units are connected to an output unit or hidden units. By using these architectures, rules can be extracted. The results of computer simulations show that this approach can generate obvious, network architectures and as a result simple rules.

Original languageEnglish
PagesI-1-I-6
Publication statusPublished - 2000 Dec 1
Externally publishedYes
Event2000 TENCON Proceedings - Kuala Lumpur, Malaysia
Duration: 2000 Sep 242000 Sep 27

Other

Other2000 TENCON Proceedings
CityKuala Lumpur, Malaysia
Period00/9/2400/9/27

    Fingerprint

ASJC Scopus subject areas

  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

Fukumi, M., Mitsukura, Y., & Akamatsu, N. (2000). New genetic approach to universal rule generation from trained neural networks. I-1-I-6. Paper presented at 2000 TENCON Proceedings, Kuala Lumpur, Malaysia, .