Neuro-fuzzy systems and their applications

Computational intelligence combines neural networks, fuzzy systems and evolutional computing. Neural networks and fuzzy systems have already proved their usefulness and have been found useful for many practical applications. We are at the beginning of the third technological revolution. Now neural networks are being capable of replacing highly skilled people with all their experience. The concept of artificial neural networks is presented, underlining their unique features and limitations. A review and comparison of various supervised and unsupervised learning algorithms follows. Several special, easy to train, architectures are shown. The neural network presentation is illustrated with many practical applications such as speaker identification, sound recognition of various equipment as a diagnosis tool, written character recognition, data compression using pulse coupled neural networks, time series prediction, etc. In the latter part of the presentation, the concept of fuzzy systems, including the conventional Zadeh approach and Takagi-Sugano architecture, is presented. The basic building blocks of fuzzy systems are discussed. Comparisons of fuzzy and neural systems are given and illustrated with several applications. The fuzzy system presentation is concluded with a description of the fabricated VLSI fuzzy chip.

[1]  B Kosko,et al.  Adaptive bidirectional associative memories. , 1987, Applied optics.

[2]  Okyay Kaynak,et al.  Neuro-fuzzy architecture for CMOS implementation , 1999, IEEE Trans. Ind. Electron..

[3]  Yoh-Han Pao,et al.  Adaptive pattern recognition and neural networks , 1989 .

[4]  Bogdan M. Wilamowski,et al.  Cmos Implementation of a voltage-Mode Fuzzy Min-Max Controller , 1996, J. Circuits Syst. Comput..

[5]  J. Hopfield.  Neurons withgraded response havecollective computational properties likethoseoftwo-state neurons , 1984 .

[6]  Jacek M. Zurada,et al.  Introduction to artificial neural systems , 1992 .

[7]  M. Sugeno,et al.  Derivation of Fuzzy Control Rules from Human Operator's Control Actions , 1983 .

[8]  Matthias M. Bauer,et al.  General Regression Neural Network for Technical Use , 1995 .

[9]  BART KOSKO,et al.  Bidirectional associative memories , 1988, IEEE Trans. Syst. Man Cybern..

[10]  R. Hecht-Nielsen.  Counterpropagation networks. , 1987, Applied optics.

[11]  Brian D. Ripley,et al.  Pattern Recognition and Neural Networks , 1996 .

[12]  Bernard Widrow,et al.  Improving the learning speed of 2-layer neural networks by choosing initial values of the adaptive weights , 1990, 1990 IJCNN International Joint Conference on Neural Networks.

[13]  J J Hopfield,et al.  Neural networks and physical systems with emergent collective computational abilities. , 1982, Proceedings of the National Academy of Sciences of the United States of America.

[14]  John J. Hopfield,et al.  Neural networks and physical systems with emergent collective computational abilities , 1999 .

[15]  Teuvo Kohonen,et al.  The self-organizing map , 1990 .

[16]  Bogdan M. Wilamowski.  Modified EBP algorithm with instant training of the hidden layer , 1997, Proceedings of the IECON'97 23rd International Conference on Industrial Electronics, Control, and Instrumentation (Cat. No.97CH36066).

[17]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .