You are here

Maximum entropy-based optimization of artificial neural networks: An application to ATM telecommunication parameter predictions

Download pdf | Full Screen View

Date Issued:
1999
Summary:
This thesis addresses studies on cost-functions developed on the basis of maximum entropy principle, for applications in artificial neural network (ANN) optimization endeavors. The maximization of entropy refers to maximizing Shannon information pertinent to the difference in the output and the teacher value of an ANN. Apart from the Shannon format of the negative entropy formulation a set of Csiszar family functions are also considered. The error-measures obtained, via these maximum entropy formulations are adopted as cost-functions in the training and prediction schedules of a test perceptron. A comparative study is done on the performance of these cost-functions in facilitating the test network towards optimization so as to predict a standard teacher function sin (.). The study is also extended to predict a parameter (such as cell delay variation) in a practical ATM telecommunication system. Concluding remarks and scope for an extended study are also indicated.
Title: Maximum entropy-based optimization of artificial neural networks: An application to ATM telecommunication parameter predictions.
0 views
0 downloads
Name(s): Sundaram, Karthik.
Florida Atlantic University, Degree grantor
De Groff, Dolores F., Thesis advisor
Neelakanta, Perambur S., Thesis advisor
College of Engineering and Computer Science
Department of Computer and Electrical Engineering and Computer Science
Type of Resource: text
Genre: Electronic Thesis Or Dissertation
Issuance: monographic
Date Issued: 1999
Publisher: Florida Atlantic University
Place of Publication: Boca Raton, Fla.
Physical Form: application/pdf
Extent: 160 p.
Language(s): English
Summary: This thesis addresses studies on cost-functions developed on the basis of maximum entropy principle, for applications in artificial neural network (ANN) optimization endeavors. The maximization of entropy refers to maximizing Shannon information pertinent to the difference in the output and the teacher value of an ANN. Apart from the Shannon format of the negative entropy formulation a set of Csiszar family functions are also considered. The error-measures obtained, via these maximum entropy formulations are adopted as cost-functions in the training and prediction schedules of a test perceptron. A comparative study is done on the performance of these cost-functions in facilitating the test network towards optimization so as to predict a standard teacher function sin (.). The study is also extended to predict a parameter (such as cell delay variation) in a practical ATM telecommunication system. Concluding remarks and scope for an extended study are also indicated.
Identifier: 9780599218833 (isbn), 15660 (digitool), FADT15660 (IID), fau:12732 (fedora)
Collection: FAU Electronic Theses and Dissertations Collection
Note(s): College of Engineering and Computer Science
Thesis (M.S.)--Florida Atlantic University, 1999.
Subject(s): Neural network (Computer science)
Asynchronous transfer mode
Held by: Florida Atlantic University Libraries
Persistent Link to This Record: http://purl.flvc.org/fcla/dt/15660
Sublocation: Digital Library
Use and Reproduction: Copyright © is held by the author, with permission granted to Florida Atlantic University to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Use and Reproduction: http://rightsstatements.org/vocab/InC/1.0/
Host Institution: FAU
Is Part of Series: Florida Atlantic University Digital Library Collections.