Detail publikace

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

BIDLO, M. VAŠÍČEK, Z.

Originální název

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

Typ

článek ve sborníku ve WoS nebo Scopus

Jazyk

angličtina

Originální abstrakt

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.

Klíčová slova

cartesian genetic programming, pattern classifier, handwritten digit recognition

Autoři

BIDLO, M.; VAŠÍČEK, Z.

Rok RIV

2014

Vydáno

21. 8. 2014

Nakladatel

Springer Verlag

Místo

Berlin

ISBN

978-3-662-44302-6

Kniha

Genetic Programming, 17th European Conference, EuroGP 2014

Edice

Lecture Notes in Computer Science

Strany od

234

Strany do

245

Strany počet

12

URL

BibTex

@inproceedings{BUT111514,
  author="Michal {Bidlo} and Zdeněk {Vašíček}",
  title="On Evolution of Multi-Category Pattern Classifiers Suitable for Embedded Systems",
  booktitle="Genetic Programming, 17th European Conference, EuroGP 2014",
  year="2014",
  series="Lecture Notes in Computer Science",
  volume="8599",
  pages="234--245",
  publisher="Springer Verlag",
  address="Berlin",
  doi="10.1007/978-3-662-44303-3\{_}20",
  isbn="978-3-662-44302-6",
  url="http://link.springer.com/chapter/10.1007/978-3-662-44303-3_20"
}