Journal of the Institute of Electronics Engineers of Korea CI (전자공학회논문지CI)

The Institute of Electronics and Information Engineers (대한전자공학회)
권호 표지

An On-line Construction of Generalized RBF Networks for System Modeling

시스템 모델링을 위한 일반화된 RBF 신경회로망의 온라인 구성

  • Kwon, Oh-Shin (School of Electrical Eng., Kunsan National Univ) ;
  • Kim, Hyong-Suk (School of Electrical Eng and Mechatronics Research Center, Chonbuk National Univ) ;
  • Choi, Jong-Soo (Research Institute of Industrial Science and Technology)
  • 권오신 (群山大學敎 電氣電子制御工學部) ;
  • 김형석 (全北大學敎 情報電子工學部 및 메카트로닉스연구센터) ;
  • 최종수 ((前)浦項産業科學院)
  • Published : 2000.01.01
Download PDF
⟨ Previous Next ⟩

Abstract

This paper presents an on-line learning algorithm for sequential construction of generalized radial basis function networks (GRBFNs) to model nonlinear systems from empirical data. The GRBFN, an extended from of standard radial basis function (RBF) networks with constant weights, is an architecture capable of representing nonlinear systems by smoothly integrating local linear models. The proposed learning algorithm has a two-stage learning scheme that performs both structure learning and parameter learning. The structure learning stage constructs the GRBFN model using two construction criteria, based on both training error criterion and Mahalanobis distance criterion, to assign new hidden units and the linear local models for given empirical training data. In the parameter learning stage the network parameters are updated using the gradient descent rule. To evaluate the modeling performance of the proposed algorithm, simulations and their results applied to two well-known benchmarks are discussed.

이 논문에서는 비선형 시스템 모델링을 위한 일반화된 RBF 신경회로망(GRBFN)을 순차적으로 구성하기 위한 온라인 학습 알고리즘을 제안한다. 상수 연결강도를 갖는 표준 RBF 신경회로망의 확장형인GRBFN은 여러 개의 국부 선형모델을 결합하여 비선형 시스템을 표현할 수 있는 구조이다. 제안한 학습 알고리즘은 구조 학습과 파라미터 학습을 수행하는 두 단계의 학습으로 구성된다. 구조 학습은 주어진 훈련 데이터로부터 새로운 은닉 유니트 및 선형 국부모델을 할항하기 위하여 훈련 오차와 Mahalanobis 거리에 기초한 두 개의 생성 조건을 이용하여 GRBFN 모델을 구성한다. 파라미터 학습은 경사강하 법칙을 기반으로 기존 네트웍의 파라미터 벡터를 갱신한다. 제안한 알고리즘의 모델링 성능을 평가하기 위해서 잘 알려진 두 예제에 대한 시뮬레이션 및 결과를 제시한다.

Keywords

References

  1. D. S. Broomhead and D. Lowe, 'Multivariate functional interpolation and adaptive networks,' Complex Systems, vol. 2, pp. 321-355, 1988
  2. M. J. D. Powell, 'Radial basis functions for multi variable interpolation: A review,' in IMA Conf. Approximation Functions Data, pp. 143-167, Shrivenham, UK, 1985
  3. J. Park and J. W. Sandberg, 'Universal approximation using radial-basis-function networks,' Neural Computation, vol. 3, pp. 246-257, 1991
  4. T. Poggio and F. Girosi, 'Regularization algorithms for learning that are equivalent to multilayer networks,' Science, vol. 247, pp, 978-982 https://doi.org/10.1126/science.247.4945.978
  5. J. Moody and C. Darken, 'Fast learning in networks of locally-tuned processing units,' Neural Computation, vol. 1, pp.281-294, 1989
  6. J. Platt, 'A resource-allocating network for function approximation,' Neural Computation, vol. 3, pp. 213-225, 1991
  7. V. Kadirkamanathan and M. Niranjan, 'A function estimation approach to sequential learning with neural networks,' Neural Computation, vol. 5, pp. 954-975, 1993
  8. L. I. Kuncheva, 'Initializing of an RBF network by a genetic algorithm,' Neurocomputing, vol. 14, pp.273-288, 1997 https://doi.org/10.1016/S0925-2312(96)00035-5
  9. R. Langari, L. Wang, and J. Yen, 'Radial basis function networks, regression weights, and the expectation-maximization algorithm,' IEEE Transactions on Systems, Man, and Cybernetics - Part A, vol. 27, no. 5, pp. 613-622, 1997 https://doi.org/10.1109/3468.618260
  10. R. Murray-Smith and K. J. Hunt, 'Local model architecture for nonlinear modelling and control,' in 'Advances in Neural Networks for Control Systems' (K. J. Hunt, G. R. Irwin, and K. Warwick, eds.), Advances in Industrial Control, Springer-Verlag, 1995
  11. K. J. Hunt, H. Haas, and R. Murray-Smith, 'Expanding the functional equivalence of radial basis function networks and fuzzy inference systems,' IEEE Transactions on Neural Networks, vol. 7, no. 3, pp.776-781, 1996 https://doi.org/10.1109/72.501735
  12. S. Chen and S. A. Billings, 'Neural networks for nonlinear dynamic system modelling and identification,' International Journal of Control, vol. 56, no. 2, pp, 319-346, 1992
  13. J. Schurmann, Pattern classification: A unified view of statistical and neural approaches, John Wiley & Sons, Inc., 1996
  14. H. -J. Zimmerman and P. Zysno, 'Latent connectives in human decision making,' Fuzzy Sets and Systems, vol. 4, pp.37-51, 1980 https://doi.org/10.1016/0165-0114(80)90062-7
  15. H. Dyckhoff and W. Pedrycz, 'Generalized means as model of compensative connectives,' Fuzzy Sets and Systems, vol. 14, pp. 143-154, 1984 https://doi.org/10.1016/0165-0114(84)90097-6
  16. R. Krishnapurum and J. Lee, 'Fuzzy-connective-based hierarchical aggregation networks for decision making,' Fuzzy Sets and Systems, vol. 46, pp. 11-27, 1992 https://doi.org/10.1016/0165-0114(92)90263-4
  17. R. Langari, and L. Wang, 'Fuzzy models, modular networks, and hybrid learning,' Fuzzy Sets and Systems, vol. 79, pp. 141-150, 1996 https://doi.org/10.1016/0165-0114(95)00151-4
  18. M. Delgado, A. F. Gomez-Skarmeta, and F. Matin, 'A fuzzy clustering-based rapid prototyping for fuzzy rule-based modeling,' IEEE Tr. on Fuzzy Systems, vol. 5, no. 2, pp. 223-233, 1997 https://doi.org/10.1109/91.580797
  19. R. M. Tong, 'The evaluation of fuzzy models derived from experimental data,' Fuzzy Sets and Systems, vol. 4, pp. 1-12, 1980 https://doi.org/10.1016/0165-0114(80)90059-7
  20. W. Pedrycz, 'An identification algorithms in fuzzy relational systems', Fuzzy Sets and Systems, vol. 13, pp. 153-167, 1984 https://doi.org/10.1016/0165-0114(84)90015-0
  21. M. Sugeno and K. Tanaka, 'Successive identification of a fuzzy model and its application to prediction of a complex system,' Fuzzy Sets and Systems, vol. 42, pp.315-334, 1991 https://doi.org/10.1016/0165-0114(91)90110-C
  22. M. Sugeno and T. Yasukawa, 'A fuzzy-logic-based approach to qualitative modeling', IEEE Trans. Fuzzy Syst., Vol. I, No.1, pp. 7-31, 1993 https://doi.org/10.1109/TFUZZ.1993.390281
  23. L. Wang and R. Langari, 'Complex systems modeling via fuzzy logic,' IEEE Tr. on Systems, Man and Cybernetics, vol. 26, No. 1, pp.100-106, 1996 https://doi.org/10.1109/3477.484441
  24. W.A. Farag, V. H. Quintana, and G. Lambert-Torres, 'A genetic-based neuro-fuzzy approach for modeling and control of dynamic systems,' IEEE Tr. on Neural Networks, Vol. 9, No.5, pp.756-767, 1998 https://doi.org/10.1109/72.712150