Publication Details

On Evolution of Multi-Category Pattern Classifiers Suitable for Embedded Systems

BIDLO Michal and VAŠÍČEK Zdeněk. On Evolution of Multi-Category Pattern Classifiers Suitable for Embedded Systems. In: Genetic Programming, 17th European Conference, EuroGP 2014. Lecture Notes in Computer Science, vol. 8599. Berlin: Springer Verlag, 2014, pp. 234-245. ISBN 978-3-662-44302-6. Available from: http://link.springer.com/chapter/10.1007/978-3-662-44303-3_20
Czech title
Evoluční návrh klasifikátorů vhodných pro vestavěné systémy
Type
conference paper
Language
english
Authors
URL
Keywords

cartesian genetic programming, pattern classifier, handwritten digit recognition

Abstract

This paper addresses the problem of evolutionary design of classifiers for the recognition of handwritten digit symbols by means of Cartesian Genetic Programming. Two different design scenarios are investigated - the design of multiple-output classifier, and design of multiple binary classifiers. The goal is to evolve classification algorithms that employ substantially smaller amount of operations in contrast with conventional approaches such as Support Vector Machines. Even if the evolved classifiers do not reach the accuracy of the tuned SVM classifier, it will be shown that the accuracy is higher than 93% and the number of required operations is a magnitude lower.

Published
2014
Pages
234-245
Proceedings
Genetic Programming, 17th European Conference, EuroGP 2014
Series
Lecture Notes in Computer Science
Volume
8599
Conference
17th European Conference on Genetic Programming, Granada, ES
ISBN
978-3-662-44302-6
Publisher
Springer Verlag
Place
Berlin, DE
DOI
EID Scopus
BibTeX
@INPROCEEDINGS{FITPUB10482,
   author = "Michal Bidlo and Zden\v{e}k Va\v{s}\'{i}\v{c}ek",
   title = "On Evolution of Multi-Category Pattern Classifiers Suitable for Embedded Systems",
   pages = "234--245",
   booktitle = "Genetic Programming, 17th European Conference, EuroGP 2014",
   series = "Lecture Notes in Computer Science",
   volume = 8599,
   year = 2014,
   location = "Berlin, DE",
   publisher = "Springer Verlag",
   ISBN = "978-3-662-44302-6",
   doi = "10.1007/978-3-662-44303-3\_20",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/10482"
}
Back to top