Beta neuro-fuzzy systems

• Autorzy: Alimi A. M.
• Języki publikacji: EN

Abstrakty

EN

In this paper we present the Beta function and its main properties. A key feature of the Beta function, which is given by the central-limit theorem, is also given. We then introduce a new category of neural networks based on a new kernel: the Beta function. Next, we investigate the use of Beta fuzzy basis functions for the design of fuzzy logic systems. The functional equivalence between Beta-based function neural networks and Beta fuzzy logic systems is then shown with the introduction of Beta neuro-fuzzy systems. By using the SW theorem and expanding the output of the Beta neuro-fuzzy system into a series of Beta fuzzy-based functions, we prove that one can uniformly approximate any real continuous function on a compact set to any arbitrary accuracy. Finally, a learning algorithm of the Beta neuro-fuzzy system is described and illustrated with numerical examples.

Słowa kluczowe

EN

beta function; kernel based neural networks; Sugeno fuzzy model; neuro-fuzzy system; universal approximation property; learning algorithms; incremental learning

• Wydawca: Politechnika Gdańska
• Czasopismo: TASK Quarterly : scientific bulletin of Academic Computer Centre in Gdansk
• Rocznik: 2003
• Tom: Vol. 7, No 1
• Strony: 23--41
• Opis fizyczny: Bibliogr. 46 poz., rys., tab.

Twórcy

autor

Alimi A. M.

• Research Group on Intelligent Machines, University of Sfax, National School of Engineers ENIS, Department of Electrical Engineering, B.P. W–3038 Sfax, Tunisia

Bibliografia

• [1] Takagi H 1990 Proc. Int. Conf. Fuzzy Logic and Neural Networks, New Orleans, USA, pp. 13–26
• [2] Wang L X 1994 Adaptive Fuzzy Systems and Control, Prentice-Hall, Englewood Cliffs, NJ
• [3] Johnson N I 1970 Continuous Univariate Distributions, Houghton Mifflin Company, Boston
• [4] Alimi A M 1998 Proc. IEEE/IMACS Multiconference on Computational EngineeringinSystems Applications: CESA’98, Hammamet, Tunisia2339
• [5] Alimi A M 1997 Proc. S ́eminaire sur la Commande Robuste et Applications: SCRA’97,Nabeul, Tunisia, pp. C1–C5
• [6] Alimi A M 2000 Int. J. ManagementInvited Paper no. 15–19
• [7] Gnedenko B V and Kolmogorov A 1954 Limit Distributions for Sums of Independent RandomVariables, Addison-Wesley Publ. Co., Reading, MA
• [8] Uspensky J V 1937 Introduction to Mathematical Probability, McGraw-Hill Book Co., NewYork
• [9] Papoulis A 1962 The Fourier Integral and Its Applications, McGraw-Hill Book Co., New York
• [10] Haykin S 1994 Neural Networks: A Comprehensive Foundation, MacMillan, New York
• [11] Pogio T and Girosi F 1990 Proc. IEEE78(9) 1481
• [12] Kalman R E 1960 Proc. 1stInt. Congress of Automatic Control, Moscow, London: Butter-worths1481
• [13] Zadeh L A 1965 Information Control8338
• [14] Zeng X J and Singh M G 1995 IEEE Trans. Fuzzy Systems3(2) 219
• [15] Lee C C 1990 IEEE Trans. Syst., Man, and Cybern.20(2) 404
• [16] Sugeno M and Kang G T 1988 Fuzzy Sets and Systems2815
• [17] Takagi T and Sugeno M 1985 IEEE Trans. Syst., Man, and Cybern.15116
• [18] Jang J S R and Sun C T 1993 IEEE Trans. Neural Networks4(1) 156
• [19] Hunt K J, Haas R and Murray-Smith R 1996 IEEE Trans. Neural Networks7(3) 776
• [20] Stone M H 1948 Math. Mag.21167;ibid.237
• [21] Stone M H 1937 AMS Trans.41375
• [22] Wang L X and Mendel J M 1992 IEEE Trans. Neural Networks3(5) 807
• [23] Alimi A M, Hassine R and Selmi M 2000 Int. J. Appl. Math. Comput. Sci.10(4) 101
• [24] Alimi A M, Hassine R and Selmi M 2002 Int. J. Appl. Math. Comput. Sci.(accepted)
• [25] Aouiti C, Alimi A M and Maalej A 2002 Systems Analysis, Modeling, and Simulation, SpecialIssue on “Advances in Control and Computer Engineering”(in print)
• [26] Aouiti C, Alimi A M and Maalej A 2000 Proc. Int. Conf. Computational and ArtificialIntelligence for Decision, Control and Automation: ACIDCA’2000, Monastir, TunisiaIM88
• [27] Hassine R, Alimi A M and Selmi M 2000 Proc. Int. Conf. Computational and ArtificialIntelligence for Decision, Control and Automation: ACIDCA’2000, Monastir, TunisiaIM72
• [28] Njah M, Alimi A M, Chtourou M and Tourki R 2002 NeuroComputing(accepted)
• [29] Njah M, Alimi A M and Chtourou M 2000 Proc. Int. Conf. Computational and ArtificialIntelligence for Decision, Control and Automation: ACIDCA’2000, Monastir, TunisiaIM76
• [30] Cho K B and Wang B H 1996 Fuzzy Sets and Systems83325
• [31] Lee S and Kil R M 1991 Neural Networks4207
• [32] Press W H, Flannery B P, Teukolsky S A and Vetterling W T 1989 Numerical Recipes in C,Cambridge University Press, Cambridge, England
• [33] Mitaim S and Kosko B 1996 Proc. IEEE Int. Conf. Fuzzy Systems, New Orleans, pp. 1237–1243
• [34] Hassine R, Alimi A M and Selmi M 2000 New Frontiers in Computational Intelligence andits Applications(Mohammadian M, Ed.), IOS Press, The Netherlands, pp. 62–67
• [35] Masmoudi M, Samet M and Alimi A M 2000 Int. J. Electronics87(6) 675
• [36] Samet M, Masmoudi M and Alimi A M 2001 Int. J. Electronics88(6) 645
• [37] Derbel N and Alimi A M 199 7Proc. Journ ́ees Tunisiennes d’Electrotechnique et d’Automa-tique: JTEA’97, Nabeul, Tunisia166
• [38] Bezine H, Derbel N and Alimi A M 2002 Engineering Applications of Artificial Intelligence(accepted)
• [39] Bezine H, Derbel N and Alimi A M 2000 Proc. Int. Conf. Computational and ArtificialIntelligence for Decision, Control and Automation: ACIDCA’2000, Monastir, TunisiaSAAC135
• [40] Ouezri A, Derbel N and Alimi A M 2002 Systems Analysis, Modeling, and Simulation, SpecialIssue on “Advances in Control and Computer Engineering”(in print)
• [41] Alimi A M and Ben Jemaa M 2002 J. Control and Intelligent Systems30(2) 47
• [42] Charfi M and Alimi A M 1998 Proc. IEEE/IMACS Multiconference on Computational Engineering in Systems Applications: CESA’98, Hammamet, Tunisia4439
• [43] Alimi A M 1998 Proc. IEEE/IMACS Multiconference on Computational Engineeringin Systems Applications: CESA’98, Hammamet, Tunisia2335
• [44] Alimi A M 1997 Proc. Int. Conf. Neural Networks: ICNN’97, Houston, TX, USA, pp. 1397–1400
• [45] Alimi A M 2002 IETE J. Research, Special Issue on “Evolutionary Computation in Engineering Sciences”(Pal S Ket al., Eds.) (in print)
• [46] Alimi A M 1997 Proc. Int. Conf. Document Analysis and Recognition: ICDAR’97, Ulm,Germany, pp. 382–386