### Abstract

The paper introduces a novel method for constructing Multilayer Perceptron (MLP) Neural Networks (NN) with the aid of fuzzy systems, particularly by deploying fuzzy J-K flip-flops as neurons. The next state Q(t+1) of the J-K fuzzy flip-flops (F^{3}) in terms of input J can be characterized by a more or less S-shaped function, for each F^{3} derived from the Yager, Dombi, and Fodor norms and co-norms. In this approach, J represents the neuron input. The other input K is wired to the complemental output (K=1-Q), thus an elementary fuzzy sequential unit with a single input and a single output is received. The algebraic F^{3} having linear J-Q(t+1) characteristics is added to the above three. The paper proposes the investigation of the possibility of constructing multilayer perceptrons from such real fuzzy hardware units. Each of the four candidates for F^{3}-based neurons is examined for its training capability by evaluating and comparing the approximation capabilities for two different transcendental functions. Simulation results are presented.

Original language | English |
---|---|

Title of host publication | IEEE International Conference on Fuzzy Systems |

Pages | 1683-1688 |

Number of pages | 6 |

DOIs | |

Publication status | Published - 2008 |

Event | 2008 IEEE International Conference on Fuzzy Systems, FUZZ 2008 - Hong Kong, China Duration: Jun 1 2008 → Jun 6 2008 |

### Other

Other | 2008 IEEE International Conference on Fuzzy Systems, FUZZ 2008 |
---|---|

Country | China |

City | Hong Kong |

Period | 6/1/08 → 6/6/08 |

### Fingerprint

### ASJC Scopus subject areas

- Software
- Artificial Intelligence
- Applied Mathematics
- Theoretical Computer Science

### Cite this

*IEEE International Conference on Fuzzy Systems*(pp. 1683-1688). [4630597] https://doi.org/10.1109/FUZZY.2008.4630597

**Multilayer perceptron implemented by fuzzy flip-flops.** / Lovassy, Rita; Kóczy, L.; Gál, László.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

*IEEE International Conference on Fuzzy Systems.*, 4630597, pp. 1683-1688, 2008 IEEE International Conference on Fuzzy Systems, FUZZ 2008, Hong Kong, China, 6/1/08. https://doi.org/10.1109/FUZZY.2008.4630597

}

TY - GEN

T1 - Multilayer perceptron implemented by fuzzy flip-flops

AU - Lovassy, Rita

AU - Kóczy, L.

AU - Gál, László

PY - 2008

Y1 - 2008

N2 - The paper introduces a novel method for constructing Multilayer Perceptron (MLP) Neural Networks (NN) with the aid of fuzzy systems, particularly by deploying fuzzy J-K flip-flops as neurons. The next state Q(t+1) of the J-K fuzzy flip-flops (F3) in terms of input J can be characterized by a more or less S-shaped function, for each F3 derived from the Yager, Dombi, and Fodor norms and co-norms. In this approach, J represents the neuron input. The other input K is wired to the complemental output (K=1-Q), thus an elementary fuzzy sequential unit with a single input and a single output is received. The algebraic F3 having linear J-Q(t+1) characteristics is added to the above three. The paper proposes the investigation of the possibility of constructing multilayer perceptrons from such real fuzzy hardware units. Each of the four candidates for F3-based neurons is examined for its training capability by evaluating and comparing the approximation capabilities for two different transcendental functions. Simulation results are presented.

AB - The paper introduces a novel method for constructing Multilayer Perceptron (MLP) Neural Networks (NN) with the aid of fuzzy systems, particularly by deploying fuzzy J-K flip-flops as neurons. The next state Q(t+1) of the J-K fuzzy flip-flops (F3) in terms of input J can be characterized by a more or less S-shaped function, for each F3 derived from the Yager, Dombi, and Fodor norms and co-norms. In this approach, J represents the neuron input. The other input K is wired to the complemental output (K=1-Q), thus an elementary fuzzy sequential unit with a single input and a single output is received. The algebraic F3 having linear J-Q(t+1) characteristics is added to the above three. The paper proposes the investigation of the possibility of constructing multilayer perceptrons from such real fuzzy hardware units. Each of the four candidates for F3-based neurons is examined for its training capability by evaluating and comparing the approximation capabilities for two different transcendental functions. Simulation results are presented.

UR - http://www.scopus.com/inward/record.url?scp=55249101975&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=55249101975&partnerID=8YFLogxK

U2 - 10.1109/FUZZY.2008.4630597

DO - 10.1109/FUZZY.2008.4630597

M3 - Conference contribution

AN - SCOPUS:55249101975

SN - 9781424418190

SP - 1683

EP - 1688

BT - IEEE International Conference on Fuzzy Systems

ER -