Performance of the Bayesian Online Algorithm for the Perceptron

Author

de Oliveira, E.A. ; Alamino, R.C.

Author_Institution

Sao Paulo Univ.

Volume

18

Issue

3

fYear

2007

fDate

5/1/2007 12:00:00 AM

Firstpage

902

Lastpage

905

Abstract

In this letter, we derive continuum equations for the generalization error of the Bayesian online algorithm (BOnA) for the one-layer perceptron with a spherical covariance matrix using the Rosenblatt potential and show, by numerical calculations, that the asymptotic performance of the algorithm is the same as the one for the optimal algorithm found by means of variational methods with the added advantage that the BOnA does not use any inaccessible information during learning

Keywords

Bayes methods; covariance matrices; perceptrons; Bayesian online algorithm; Rosenblatt potential; continuum equations; generalization error; one-layer perceptron; spherical covariance matrix; variational methods; Algorithm design and analysis; Approximation algorithms; Bayesian methods; Covariance matrix; Equations; Gradient methods; Machine learning; Machine learning algorithms; Parameter estimation; Pattern classification; Bayesian algorithms; online gradient methods; pattern classification; Algorithms; Artificial Intelligence; Bayes Theorem; Computer Simulation; Decision Support Techniques; Information Storage and Retrieval; Models, Theoretical; Neural Networks (Computer); Online Systems; Pattern Recognition, Automated;

fLanguage

English

Journal_Title

Neural Networks, IEEE Transactions on

Publisher

ieee

ISSN

1045-9227

Type

jour

DOI

10.1109/TNN.2007.891189

Filename

4182376