Title

Supervised Competitive Learning Part I: SCL with Backpropagation Networks

Authors

Thomas H. Fuller Jr., Washington University in St Louis
Takayuki D. Kimura, Washington University in St Louis

Document Type

Technical Report

Publication Date

1993

Filename

WUCS-93-43.pdf

DOI:

10.7936/K7BP014F

Technical Report Number

WUCS-93-43

Abstract

SCL assembles a set of learning modules into a supervised learning system to address the stability-plasticity dilemma. Each learning module acts as a similarity detector for a prototype, and includes prototype resetting (akin to that of ART) to respond to new prototypes. Here (Part I) we report SCL results using back-propagation networks as the learning modules. We used two feature extractors: about 30 energy-based features, and a combination of energy-based and graphical features (about 60). SCL recognized 98% (energy) and 99% (energy/graphical) of test digits, and 91% (energy) and 96% (energy/graphical) of test letters. In the accompanying paper (Part II), we report the results of SCL using fuzzy sets as learning moduels for recognizing handwritten digits.

Comments

Permanent URL: http://dx.doi.org/10.7936/K7BP014F

Recommended Citation

Fuller, Thomas H. Jr. and Kimura, Takayuki D., "Supervised Competitive Learning Part I: SCL with Backpropagation Networks" Report Number: WUCS-93-43 (1993). All Computer Science and Engineering Research.
https://openscholarship.wustl.edu/cse_research/538