• 検索結果がありません。

ファジィ差分進化手法によるファジィニューラルネットの進化的学習

N/A
N/A
Protected

Academic year: 2021

シェア "ファジィ差分進化手法によるファジィニューラルネットの進化的学習"

Copied!
2
0
0

読み込み中.... (全文を見る)

全文

(1)

B-06

Evolving Fuzzy Neural Networks by means of Fuzzy Differential Evolution

Hidehiko Okada

1

genotype [1]

1 Fuzzy

Differential Evolution: FDE

(FNN)[2]

2

[3] [2] FNN FNN 3 FNN 1 FNN 1 1: [2] FNN x O (1)-(5) (1)-(5) , , , , , , , , [2] 2 c w L U [3] 2: [3] 1 FNN [2]

3

DE

FNN FNN [3] DE genotype genotype Xi = [xiL, xiU] Xi = (xic, xiw) xiL, xiU, xic, xiw Xi [3] LU CW FDE DE[4] genotype DE DE/rand/1/bin

genotype Xa Xa Xa Xa donor vector

trial vector Ya Za Ya 3 Xb1 Xb2 Xb3 Xb1 Xb2 Xb3 DE/rand/1/bin Xa Xb1 Xb2 Xb3 Xa Xa Ya Za Xb1 Xb2 Xb3 LU , Faculty of Computer Science and Engineering, Kyoto Sangyo University

(2)

(6),(7) ( CW (9) (6)-(9) F DE/rand/1/bin DE/rand/1/bin DE/rand/1/bin FDE DE donor vector DE/rand/1/bin

4

[5] [2] FNN FDE 3 [3] 3 F0.0L F0.0U F1.0 F(x) [3] FNN =10 FDE =100 =10,000 F=0.5 CR=0.8 genotype =LU FNN FNN 4 FNN 5 4 5 FNN 3

5

DE NN FNN

[1] H. Okada, Proposal of fuzzy evolutionary algorithms with

fuzzy valued genotypes, Proc. of International Conference on Instrumentation, Control and Information Technology (SICE Annual Conference 2012), 1538-1541, 2012. [2] H. Ishibuchi, H. Tanaka and H. Okada, Fuzzy neural

networks with fuzzy weights and fuzzy biases, Proc. of IEEE International Conferences on Neural Networks, 1650-1655, 1993.

[3] , GA

,

25 , C-5, 2013.

[4] R. Storn and K. Price, Differential evolution - a simple and efficient heuristic for global optimization over continuous spaces, Journal of Global Optimization,11(4), 341-359, 1997.

[5] X. Yao, Evolving artificial neural networks, Proc. of the IEEE, 87(9), 1423-1447, 1999.

3: [3]

4: FNN

参照

関連したドキュメント

以上,本研究で対象とする比較的空気を多く 含む湿り蒸気の熱・物質移動の促進において,こ

Optimal stochastic approximation algorithms for strongly convex stochastic composite optimization I: A generic algorithmic framework.. SIAM Journal on Optimization,

The performance of scheduling algorithms for LSDS control is usually estimated using a certain number of standard parameters, like total time or schedule

We present the new multiresolution network flow minimum cut algorithm, which is es- pecially efficient in identification of the maximum a posteriori (MAP) estimates of corrupted

We present the new multiresolution network flow minimum cut algorithm, which is es- pecially efficient in identification of the maximum a posteriori (MAP) estimates of corrupted

The complexity of dynamic languages and dynamic optimization problems. Lipschitz continuous ordinary differential equations are

 少子高齢化,地球温暖化,医療技術の進歩,AI

The geometric configurations of singularities at infinity of the family of quadratic systems possessing finite singularities of total multiplicity 4 (i.e. µ 0 6= 0) are classified