Publication detail

Optimization of multilayer perceptron training parameters using artificial bee colony and genetic algorithm

KARTCI, A.

Original Title

Optimization of multilayer perceptron training parameters using artificial bee colony and genetic algorithm

Czech Title

Optimalizace parametrů školení vícevrstvého perceptronu použitím umělého včelstva a genetického algoritmu

English Title

Optimization of multilayer perceptron training parameters using artificial bee colony and genetic algorithm

Type

conference paper

Language

en

Original Abstract

In this paper, the momentum coefficient, learning rate, and the number of hidden neurons where the multilayer perceptron works best, are determined. The network and optimization algorithms are written in MATLAB, which was also successfully used to carry out results. To obtain the results, IRIS, mammographic_mass, and new_thyroid data sets have been used. Obtained results show that the determining effect on the neural learning process of parameters (momentum coefficient, learning rate, number of hidden neurons) are compatible with other approaches available in the literature. Both genetic algorithm (GA) and artificial bee colony (ABC) algorithm were successful on finding the values to get high performance as well as effect on performance of the population number.

Czech abstract

In this paper, the momentum coefficient, learning rate, and the number of hidden neurons where the multilayer perceptron works best, are determined. The network and optimization algorithms are written in MATLAB, which was also successfully used to carry out results. To obtain the results, IRIS, mammographic_mass, and new_thyroid data sets have been used. Obtained results show that the determining effect on the neural learning process of parameters (momentum coefficient, learning rate, number of hidden neurons) are compatible with other approaches available in the literature. Both genetic algorithm (GA) and artificial bee colony (ABC) algorithm were successful on finding the values to get high performance as well as effect on performance of the population number.

English abstract

In this paper, the momentum coefficient, learning rate, and the number of hidden neurons where the multilayer perceptron works best, are determined. The network and optimization algorithms are written in MATLAB, which was also successfully used to carry out results. To obtain the results, IRIS, mammographic_mass, and new_thyroid data sets have been used. Obtained results show that the determining effect on the neural learning process of parameters (momentum coefficient, learning rate, number of hidden neurons) are compatible with other approaches available in the literature. Both genetic algorithm (GA) and artificial bee colony (ABC) algorithm were successful on finding the values to get high performance as well as effect on performance of the population number.

Keywords

Multilayer perceptron, artificial bee colony algorithm, genetic algorithm, training parameters optimization

RIV year

2015

Released

23.04.2015

Publisher

Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologií

Location

Brno

ISBN

978-80-214-5148-3

Book

Proceedings of the 21st Conference STUDENT EEICT 2015

Pages from

338

Pages to

340

Pages count

3

BibTex


@inproceedings{BUT117521,
  author="Aslihan {Kartci}",
  title="Optimization of multilayer perceptron training parameters using artificial bee colony and genetic algorithm",
  annote="In this paper, the momentum coefficient, learning rate, and the number of hidden neurons where the multilayer perceptron works best, are determined. The network and optimization algorithms are written in MATLAB, which was also successfully used to carry out results. To obtain the results, IRIS, mammographic_mass, and new_thyroid data sets have been used. Obtained results show that the determining effect on the neural learning process of parameters (momentum coefficient, learning rate, number of hidden neurons) are compatible with other approaches available in the literature. Both genetic algorithm (GA) and artificial bee colony (ABC) algorithm were successful on finding the values to get high performance as well as effect on performance of the population number.",
  address="Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologií",
  booktitle="Proceedings of the 21st Conference STUDENT EEICT 2015",
  chapter="117521",
  howpublished="electronic, physical medium",
  institution="Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologií",
  year="2015",
  month="april",
  pages="338--340",
  publisher="Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologií",
  type="conference paper"
}