Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/3701
Title: Approximation of a fractal curve using feed-forward neural networks
Authors: Ho, Wai-shing
Keywords: Neural networks (Computer science)
Fractals
Hong Kong Polytechnic University -- Dissertations
Issue Date: 2000
Publisher: The Hong Kong Polytechnic University
Abstract: The approximation of fractal curves in the form of Brownian functions by two-layer feed-forward neural networks is studied. The network parameters are restricted within a finite range. For given realizations of the Brownian target function, all local minima in the output error measure with appreciable sizes of basins of attraction are located and found to be about a dozen in number in each case. The error follows a log-normal distribution which can be explained by a distribution of mean squared normal deviates. Its mean value exhibits simple scaling relationships with the number of hidden neurons and the number of training patterns. Our numerical findings are explained by comparison with a simple piecewise linear fit approach.
Description: 36 leaves : ill. ; 30 cm.
PolyU Library Call No.: [THS] LG51 .H577M AP 2000 Ho
URI: http://hdl.handle.net/10397/3701
Rights: All rights reserved.
Appears in Collections:Thesis

Files in This Item:
File Description SizeFormat 
b15249141_link.htmFor PolyU Users 162 BHTMLView/Open
b15249141_ir.pdfFor All Users (Non-printable) 1.63 MBAdobe PDFView/Open
Show full item record

Page view(s)

376
Last Week
0
Last month
Checked on Sep 25, 2016

Download(s)

176
Checked on Sep 25, 2016

Google ScholarTM

Check



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.