Parameter optimisation in fuzzy flip-flop-based neural networks

Rita Lovassy, L. ászlóT Kóczy, L. ászlóT Kóczy, L. ászló Gál, L. ászló Gál

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

This paper presents a method for optimising the parameters of fuzzy flip-flop-based neural networks (FNN) consisting of fuzzy J-K and D flip-flop neurons based on various popular fuzzy operations using bacterial memetic algorithm with the modified operator execution order (BMAM). In early works, the authors proposed the Levenberg-Marquardt algorithm (LM) a widely used second order gradient type training algorithm for fuzzy neural networks variables optimisation. The BMAM local and global search evolutionary approach is a bacterial type memetic algorithm which executes several LM cycles during the bacterial mutation after each mutational step, using the LM method more efficiently. Numerical experiments were performed to show the function approximation capability of various quasi optimised FNN types based on fuzzy J-K and D flip-flop neurons using algebraic, Lukasiewicz, Yager, Dombi, Hamacher and Frank norms, trained with LM method and BMAM algorithm.

Original languageEnglish
Pages (from-to)237-243
Number of pages7
JournalInternational Journal of Reasoning-based Intelligent Systems
Volume2
Issue number3-4
DOIs
Publication statusPublished - 2010

Fingerprint

Flip flop circuits
Neural networks
Neurons
Fuzzy neural networks
Mathematical operators

Keywords

  • bacterial memetic algorithm with the modified operator execution order
  • BMAM
  • FNN
  • fuzzy D flip-flop
  • fuzzy flip-flop neural network
  • fuzzy flip-flop neuron
  • fuzzy J-K flip-flop

ASJC Scopus subject areas

  • Computer Science(all)
  • Engineering(all)

Cite this

Parameter optimisation in fuzzy flip-flop-based neural networks. / Lovassy, Rita; Kóczy, L. ászlóT; Kóczy, L. ászlóT; Gál, L. ászló; Gál, L. ászló.

In: International Journal of Reasoning-based Intelligent Systems, Vol. 2, No. 3-4, 2010, p. 237-243.

Research output: Contribution to journalArticle

Lovassy, Rita ; Kóczy, L. ászlóT ; Kóczy, L. ászlóT ; Gál, L. ászló ; Gál, L. ászló. / Parameter optimisation in fuzzy flip-flop-based neural networks. In: International Journal of Reasoning-based Intelligent Systems. 2010 ; Vol. 2, No. 3-4. pp. 237-243.
@article{f6ef5f8f51c043859b711f4510e2b9d8,
title = "Parameter optimisation in fuzzy flip-flop-based neural networks",
abstract = "This paper presents a method for optimising the parameters of fuzzy flip-flop-based neural networks (FNN) consisting of fuzzy J-K and D flip-flop neurons based on various popular fuzzy operations using bacterial memetic algorithm with the modified operator execution order (BMAM). In early works, the authors proposed the Levenberg-Marquardt algorithm (LM) a widely used second order gradient type training algorithm for fuzzy neural networks variables optimisation. The BMAM local and global search evolutionary approach is a bacterial type memetic algorithm which executes several LM cycles during the bacterial mutation after each mutational step, using the LM method more efficiently. Numerical experiments were performed to show the function approximation capability of various quasi optimised FNN types based on fuzzy J-K and D flip-flop neurons using algebraic, Lukasiewicz, Yager, Dombi, Hamacher and Frank norms, trained with LM method and BMAM algorithm.",
keywords = "bacterial memetic algorithm with the modified operator execution order, BMAM, FNN, fuzzy D flip-flop, fuzzy flip-flop neural network, fuzzy flip-flop neuron, fuzzy J-K flip-flop",
author = "Rita Lovassy and K{\'o}czy, {L. {\'a}szl{\'o}T} and K{\'o}czy, {L. {\'a}szl{\'o}T} and G{\'a}l, {L. {\'a}szl{\'o}} and G{\'a}l, {L. {\'a}szl{\'o}}",
year = "2010",
doi = "10.1504/IJRIS.2010.036869",
language = "English",
volume = "2",
pages = "237--243",
journal = "International Journal of Reasoning-based Intelligent Systems",
issn = "1755-0556",
publisher = "Inderscience Publishers",
number = "3-4",

}

TY - JOUR

T1 - Parameter optimisation in fuzzy flip-flop-based neural networks

AU - Lovassy, Rita

AU - Kóczy, L. ászlóT

AU - Kóczy, L. ászlóT

AU - Gál, L. ászló

AU - Gál, L. ászló

PY - 2010

Y1 - 2010

N2 - This paper presents a method for optimising the parameters of fuzzy flip-flop-based neural networks (FNN) consisting of fuzzy J-K and D flip-flop neurons based on various popular fuzzy operations using bacterial memetic algorithm with the modified operator execution order (BMAM). In early works, the authors proposed the Levenberg-Marquardt algorithm (LM) a widely used second order gradient type training algorithm for fuzzy neural networks variables optimisation. The BMAM local and global search evolutionary approach is a bacterial type memetic algorithm which executes several LM cycles during the bacterial mutation after each mutational step, using the LM method more efficiently. Numerical experiments were performed to show the function approximation capability of various quasi optimised FNN types based on fuzzy J-K and D flip-flop neurons using algebraic, Lukasiewicz, Yager, Dombi, Hamacher and Frank norms, trained with LM method and BMAM algorithm.

AB - This paper presents a method for optimising the parameters of fuzzy flip-flop-based neural networks (FNN) consisting of fuzzy J-K and D flip-flop neurons based on various popular fuzzy operations using bacterial memetic algorithm with the modified operator execution order (BMAM). In early works, the authors proposed the Levenberg-Marquardt algorithm (LM) a widely used second order gradient type training algorithm for fuzzy neural networks variables optimisation. The BMAM local and global search evolutionary approach is a bacterial type memetic algorithm which executes several LM cycles during the bacterial mutation after each mutational step, using the LM method more efficiently. Numerical experiments were performed to show the function approximation capability of various quasi optimised FNN types based on fuzzy J-K and D flip-flop neurons using algebraic, Lukasiewicz, Yager, Dombi, Hamacher and Frank norms, trained with LM method and BMAM algorithm.

KW - bacterial memetic algorithm with the modified operator execution order

KW - BMAM

KW - FNN

KW - fuzzy D flip-flop

KW - fuzzy flip-flop neural network

KW - fuzzy flip-flop neuron

KW - fuzzy J-K flip-flop

UR - http://www.scopus.com/inward/record.url?scp=84952969047&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84952969047&partnerID=8YFLogxK

U2 - 10.1504/IJRIS.2010.036869

DO - 10.1504/IJRIS.2010.036869

M3 - Article

AN - SCOPUS:84952969047

VL - 2

SP - 237

EP - 243

JO - International Journal of Reasoning-based Intelligent Systems

JF - International Journal of Reasoning-based Intelligent Systems

SN - 1755-0556

IS - 3-4

ER -