IMPLEMENTATION OF BOLTZMANN MACHINE IN NEURO-SYMBOLIC INTEGRATION

Loading...
Thumbnail Image
Date
2010-04
Authors
TEOH, YEONG KIN
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In this dissertation, we present and compare two methods of doing logic program on a Hopfield network based on the energy minimization scheme. The proposed method is based on the Boltzmann Machine model. Computer simulations to demonstrate the ability of Boltzmann Machine doing logic program on a Hopfield network will be discussed. Besides that, global minima ratio of network programmed by program clauses in the features of running time and complexity are also been analyzed. Program experiments were conducted using Microsoft Basic C 6.0 software. The present study shows that the Boltzmann Machine model is more stable and efficient. Given a sufficient time, Boltzmann Machine is able to represent and solve difficult combinatory problems.
Description
Keywords
IMPLEMENTATION , SYMBOLIC
Citation