A learning algorithm for tuning fuzzy inference rules

In this paper, by using the gradient descent method we propose a tuning approach to obtain optimal fuzzy inference rules in which the membership functions are nonsymmetrical triangular-type membership functions. In the tuning approach, the representation of the fuzzy rule table does not change even after the learning which shows that it is intuitive and convenient for practical fuzzy applications. Moreover the efficiency of the presented method is also demonstrated by means of identifying nonlinear systems.

[1]  Yan Shi,et al.  Some considerations on conventional neuro-fuzzy learning algorithms by gradient descent method , 2000, Fuzzy Sets Syst..

[2]  Dimitar Filev,et al.  Generation of Fuzzy Rules by Mountain Clustering , 1994, J. Intell. Fuzzy Syst..

[3]  Jerry M. Mendel,et al.  Back-propagation fuzzy system as nonlinear dynamic system identifiers , 1992, [1992 Proceedings] IEEE International Conference on Fuzzy Systems.

[4]  Naoyoshi Yubazaki,et al.  A learning algorithm for tuning fuzzy rules based on the gradient descent method , 1996, Proceedings of IEEE 5th International Fuzzy Systems.

[5]  Yan Shi,et al.  Tuning Fuzzy Rules Using Fuzzy Singleton-Type Reasoning Method , 1997 .

[6]  M. Mizumoto,et al.  An improvement of neuro-fuzzy learning algorithm for tuning fuzzy rules based on fuzzy clustering method , 1998, 1998 IEEE International Conference on Fuzzy Systems Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98CH36228).

[7]  Isao Hayashi,et al.  A learning method of fuzzy inference rules by descent method , 1992 .