Off-line structural risk minimization and BARTMAP-S

Author

Verzi, Stephen J. ; Heileman, Gregory L. ; Georgiopoulos, Michael ; Anagnostopoulos, Georgios

Author_Institution

Dept. of Comput. Sci., New Mexico Univ., Albuquerque, NM, USA

Volume

3

fYear

2002

fDate

6/24/1905 12:00:00 AM

Firstpage

2533

Lastpage

2538

Abstract

BARTMAP-S (Simplified Boosted ARTMAP) is a neural network architecture with which structural risk minimization can be performed, although indirectly. BARTMAP-S is trained in an online fashion, consistent with the original way intended for the fuzzy ARTMAP neural network architecture. We propose an extension to BARTMAP-S for conducting off-line learning. Consequently, this alternate mode of learning will allow us to conduct structural risk minimization more directly. We describe the new architecture and present some empirical results to demonstrate the usefulness of structural risk minimization in learning with an ARTMAP-based neural network

Keywords

ART neural nets; generalisation (artificial intelligence); learning (artificial intelligence); neural net architecture; BARTMAP-S; Simplified Boosted ARTMAP; adaptive resonance theory; empirical risk minimization; generalization performance; neural network architecture; off-line learning; off-line structural risk minimization; overlapping pattern classes; Adaptive systems; Computer networks; Computer science; Fuzzy logic; Machine learning; Neural networks; Resonance; Risk management; Virtual colonoscopy;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 2002. IJCNN '02. Proceedings of the 2002 International Joint Conference on

Conference_Location

Honolulu, HI

ISSN

1098-7576

Print_ISBN

0-7803-7278-6

Type

conf

DOI

10.1109/IJCNN.2002.1007542

Filename

1007542