臺灣博碩士論文加值系統

本文主要提出教導式及增強式進化學習演算法用於遞迴式小波類神經模糊網路，在本文中遞迴式小波類神經模糊網路是以小波為基底之類神經模糊網路本身為多層網路架構，並整合了傳統TSK 模糊模組及小波類神經網路。而遞迴特性主要是源自於歸屬函數中激發程度，作為內部輸送的變數值，回傳給本身(歸屬函數)做為輸入。在學習演算方面包括教導式進化學習及增強式進化學習。在教導式學習方面，我們提出了動態共生進化與自我建構進化演算法。在動態共生進化演算法主要是提出產生較佳的初始族群以及找出較佳的突變點。而在自我建構進化演算法中我們使用一個子族群來評估取局部解以及多個子族群架構一個遞迴式小波類神經模糊網路的族群架構來取代動態共生進化演算法的族群架構。此外我們也提出了自我架構學習演算法根據輸入資料自動化架構遞迴式小波類神經模糊網路。在自我建構進化演算法的參數學習部分則很類似動態共生進化演算法的學習方式。雖然動態共生進化演算法和自我建構進化演算法在實驗結果中有很好的表現，但這兩個演算法是屬於教導式學習的應用。然而在真實世界中有些資料卻是很昂貴或是很難去取得的。為了解決這樣的問題，我們提出了一個增強式動態共生進化演算法。在增強式動態共生進化演算法中主要是利用一個紀錄錯誤發生時次數的機制作為設計適應函數的方法。在參數學習方面是使用動態共生進化演算法的學習方式。在實驗結果中我們可以看出本文所提出的教導式和增強式進化學習的效能。

In this thesis, supervised and reinforcement evolution learning methods are proposed for recurrent wavelet neuro-fuzzy networks (RWNFN). The RWNFN model is a feedforward multi-layer network which integrates traditional Takagi-Sugeno-Kang (TSK) fuzzy model and the wavelet neural networks (WNN). The recurrent property comes from feeding the internal variables, derived from membership function matched degree, back to itself. In the learning algorithm, this thesis proposed supervised and reinforcement evolution learning methods. The supervised evolution learning methods consist of the dynamic symbiotic evolution (DSE) and the self-constructing evolution algorithm (SCEA). In the DSE, the better chromosomes will be initially generated while the better mutation points will be determined for performing dynamic mutation. In the SCEA, we modified the structure of population in the DSE that we use a subpopulation to evaluate a partial solution locally and several subpopulations to construct a full solution. Moreover, the proposed SCEA uses the self-constructing learning algorithm to construct the RWNFN model automatically that is based on the input training data to decide the input partition. The DSE is using to carry out parameter learning of the RWNFN model in the SECA. Although the DSE and SCEA can obtain good performance in the simulations, however in some real-world applications exact training data may be expensive or even impossible to obtain. To solve this problem, the reinforcement evolution learning method called the reinforcement dynamic symbiotic evolution (R-DSE) is proposed. In the R-DSE, we formulate a number of time steps before failure occurs as the fitness function. The DSE is used as a way to perform parameter learning. In the simulations, efficiency of the proposed supervised and reinforcement learning methods are verified from these results.

Abstract in Chinese IV
Abstract in English VI
Contents X
List of Figures XII
List of Tables XVIII
Chapter I : INTRODUCTION 1
1.1 Motivation 1
1.2 Literature Review 4
1.3 Thesis Organization 10
Chapter II : A RECURRENT WAVELET NEURO-FUZZY NETWORKS (RWNFN) 11
2.1 Introduction 11
2.2 Wavelet Bases and Wavelet Neural Networks 13
2.3 Structure of the RWNFN Model 16
Chapter III : THE SUPERVISED EVOLUTION LEARNING METHODS
21
3.1 Introduction 21
3.2 Dynamic Symbiotic Evolution 23
3.2.1 The DSE Method 23
3.2.2 Illustrative Example 36
3.3 Self-Constructing Evolution Algorithm 60
3.3.1 The SCEA Method 60
3.3.2 Illustrative Example 71
3.4 Discussion 82
Chapter IV : THE REINFORCEMENT EVOLUTION LEARNING METHOD 84
4.1 Introduction 84
4.2 The Reinforcement Dynamic Symbiotic Evolution 85
4.2.1 The R-DSE Method 85
4.2.2 Illustrative Example 88
4.3 Discussion 112
Chapter V : CONCLUSION 113
Bibliography 115
Vita 125
Publication List 126

[1] D. E. Goldberg, Genetic Algorithms in Search Optimization and Machine Learning. Reading, MA: Addison-Wesley, 1989.
[2] J. K. Koza, Genetic Programming: On the Programming of Computers by Means of Natural Selection. Cambridge, MA: MIT Press, 1992.
[3] L. J. Fogel, “Evolutionary programming in perspective: The top-down view,” in Computational Intelligence: Imitating Life, J. M. Zurada, R. J. Marks II, and C. Goldberg, Eds. Piscataway, NJ: IEEE Press, 1994.
[4] I. Rechenberg, “Evolution strategy,” in Computational Intelligence: Imitating Life, J. M. Zurada, R. J. Marks II, and C. Goldberg, Eds. Piscataway, NJ: IEEE Press, 1994.
[5] C. L. Karr, “Design of an adaptive fuzzy logic controller using a genetic algorithm,” in Proc. 4th Conf. Genetic Algorithms, pp. 450-457, 1991.
[6] A. Homaifar and E. McCormick, “Simultaneous design of membership functions and rule sets for fuzzy controllers using genetic algorithms,” IEEE Trans. Fuzzy Syst., vol. 3, no. 9, pp. 129-139, May 1995.
[7] M. A. Lee and H. Takagi, “Integrating design stages of fuzzy systems using genetic algorithms,” in Proc. IEEE Int. Conf. Fuzzy Syst., New York, vol. 1, pp. 612-617, Apr. 1993.
[8] C. F. Juang, “A TSK-type recurrent fuzzy network for dynamic systems processing by neural network and genetic algorithms,” IEEE Trans. Fuzzy Systs., vol. 10, no. 2, pp. 155-170, Apr. 2002.
[9] C. F. Juang, J. Y. Lin and C. T. Lin, “Genetic reinforcement learning through symbiotic evolution for fuzzy controller design,” IEEE Trans. Syst., Man, Cybern., Part B, vol. 30, no. 2, pp. 290-302, Apr. 2000.
[10] C. T. Lin and C. P. Jou, “GA-based fuzzy reinforcement learning for control of a magnetic bearing system,” IEEE Trans. Syst., Man, Cybern., Part B, vol. 30, no. 2, pp. 276-289, Apr. 2000.
[11] P. Kumar, V.K. Chandna, Thomas, “Fuzzy-genetic algorithm for pre-processing data at the RTU,” IEEE Trans. Power Systems, vol. 19, no. 2, pp. 718-723, May. 2004.
[12] W. Y. Wan and Y. H. Li, “Evolutionary learning of BMF fuzzy-neural networks using a reduced-form genetic algorithm,” IEEE Trans. Syst., Man, Cybern., Part B, vol. 33, no. 6, pp. 966–976, Dec. 2003.
[13] C. T. Lin and C. S. G. Lee, Neural Fuzzy Systems: A Neuro-Fuzzy Synergism to Intelligent System, NJ:Prentice-Hall, 1996.
[14] G. G. Towell and J. W. Shavlik, “Extracting refined rules from knowledge-based neural networks,” Machine Learning, vol. 13, pp. 71-101, 1993.
[15] C. J. Lin and C. T. Lin, “An ART-based fuzzy adaptive learning control network,” IEEE Trans. Fuzzy systs., vol. 5, no. 4, pp. 477-496, Nov. 1997.
[16]L. X. Wang and J. M. Mendel, “Generating fuzzy rules by learning from examples,” IEEE Trans. Syst., Man, Cybern., vol. 22, no. 6, pp. 1414-1427, Nov/Dec. 1992.
[17]T. Takagi and M. Sugeno, “Fuzzy identification of systems and its applications to modeling and control,” IEEE Trans. Syst., Man, Cybern., vol. SMC-15, pp. 116-132, 1985.
[18]J.-S. R. Jang, ”ANFIS: Adaptive-network-based fuzzy inference system,” IEEE Trans. on Syst., Man, and Cybern., vol. 23, pp. 665-685, 1993.
[19]C. F. Juang and C. T. Lin, “An self-constructing self-constructing neural fuzzy inference network and its applications,” IEEE Trans. Fuzzy Systs., vol. 6, no.1, pp. 12-31, Feb. 1998.
[20] F. J. Lin, C. H. Lin, and P. H. Shen, “Self-constructing fuzzy neural network speed controller for permanent-magnet synchronous motor drive,” IEEE Trans. Fuzzy Systs., vol. 9, no. 5, pp. 751-759, Oct. 2001.
[21] H. Takagi, N. Suzuki, T. Koda, and Y. Kojima, “Neural networks designed on approximate reasoning architecture and their application,” IEEE Trans. Neural Networks, vol. 3, no. 5, pp. 752-759, 1992.
[22] K.J. °Astrom and B. Wittenmark, Adaptive Control. Reading, MA: Addison-Wesley, 1989.
[23] J. Hauser, S. Sastry, and P. Kokotovic, “Nonolinear control via approximate input-output linearization: the ball and beam example,” IEEE Trans. Automatic Control, vol. 37, pp. 392-398, Mar. 1992.
[24] E. Mizutani and J.-S. R. Jang, “Coactive neural fuzzy modeling,” Proc. Int. Conf. Neural Networks, pp. 760-765, 1995.
[25] C. J. Lin and C. C. Chin, “A wavelet-based neuro-fuzzy system and its applications,” Proc. IEEE Int. Conf. Neural Networks, pp. 1921-1926, Oregon, July 20-24, 2003.
[26] K. S. Narendra and K. Parthasarathy, ”Identification and control of dynamical systems using neural networks,” IEEE Trans. Neural Networks, vol. 1, pp. 4-27, 1990.
[27] J. Zhang and A. J. Morris, “Recurrent neuro-fuzzy networks for nonlinear process modeling,” IEEE Trans. Neural Networks, vol. 10, no. 2, pp.313-326, 1999.
[28] C. F. Juang and C. T. Lin, “A recurrent self-organizing neural fuzzy inference network,” IEEE Trans. Neural Networks, vol. 10, no. 4, pp.828-845, July 1999.


[29] C. H. Lee and C. C. Teng, “Identification and control of dynamic systems using recurrent fuzzy neural networks,” IEEE Trans. Fuzzy Systs., vol. 8, no. 4, pp. 349-366, Aug. 2000.
[30] P. A. Mastorocostas and J. B. Theocharis, “A recurrent fuzzy-neural model for dynamic system identification,” IEEE Trans. Syst., Man, Cybern., vol. 32, no. 2, pp. 176-190, Apr. 2002.
[31] S. F. Su and F. Y. Yang, "On the dynamical modeling with neural fuzzy networks," IEEE Trans. Neural Networks, vol. 13, no. 6, pp. 1548-1553, Nov. 2002.
[32] G. Mouzouris and J. M. Mendel, “Dynamic non-singleton fuzzy logic systems for nonlinear modeling,” IEEE Trans. Fuzzy Systs., vol.5, pp. 199-208, 1997.
[33] Q. Zhang and A. Benveniste, “Wavelet networks,” IEEE Trans. Neural Networks, Vol. 3, No. 6, Nov. 1992, pp. 889–898.
[34] Y. C. Pati and P. S. Krishnaprasad, “Analysis and synthesis of feedforward neural networks using discrete affine wavelet transformations,” IEEE Trans. on Neural Networks, Vol. 4, No. 2, 1993, pp. 73-85.

[35] H. H. Szu and S. Kadambe, “Neural network adaptive wavelets for signal representation and classification,” Optical Engineering, Vol. 31, No. 9, 1992, pp. 1907-1916.
[36] T. Yamakawa, E. Uchino and T. Samatsu, “Wavelet neural networks employing over-complete number of compactly supported non-orthogonal wavelets and their applications,” Proc. of IEEE Conf. on Neural Networks, Vol. 3, 1994, pp. 1391 –1396.
[37] J. Chen and D. D. Bruns, “WaveARX neural network development for system identification using a systematic design synthesis,” Ind. Eng. Chem. Res., Vol. 34, 1995, pp. 4420–4435.
[38] I. Daubechies, ”Orthonormal Bases of Compactly Supported Wavelets,” Comm. Pur. Appl. Math., Vol. 41, 1998.
[39] Ho D.W.C., P. A. Zhang and J. Xu, “Fuzzy Wavelet Networks for Function Learning,” IEEE Trans. on Fuzzy Systems, Vol. 9, No. 1, 2001, pp. 200 –211.
[40]T. Bäck and H. P. Schwefel, “An overview of evolutionary algorithms for parameter optimization,” Evolutionary Comput., vol. 1, no. 1, pp. 1–23, 1993.
[41]D. B. Fogel, Evolutionary Computation: Toward a New Philosophy of Machine Intelligence. Piscataway, NJ: IEEE Press, 1995.
[42]X. Yao, Ed., Evolutionary Computation: Theory and Applications, Singapore: World Scientific, 1999.
[43] R. E. Smith, S. Forrest, and A. S. Perelson, “Searching for diverse, cooperative populations with genetic algorithms,” Evol. Comput., vol. 1, no. 2, pp. 127–149, 1993.
[44] D. E. Moriarty and R. Miikkulainen, “Efficient reinforcement learning through symbiotic evolution,” Mach. Learn., vol. 22, pp. 11–32, 1996.
[45] J.-S. R. Jang, C. T. Sun, and E. Mizutani, Neuro-Fuzzy and Soft Computing, Ch. 17, Prentice-Hall, 1997.
[46] O. Cordon, F. Herrera, F. Hoffmann, and L. Magdalena, Genetic fuzzy systems evolutionary tuning and learning of fuzzy knowledge bases. Advances in Fuzzy Systems-Applications and Theory, vol.19, NJ: World Scientific Publishing, 2001
[47] J. Tanomaru and S. Omatu, “Process control by self-constructing trained neural controllers,” IEEE Trans. on Ind. Electron., Vol. 39, pp. 511-521, 1992.

[48] R. J. Williams and D. Zipser, “A learning algorithm for continually running recurrent neural networks,” Neural Comput., vol. 1, no. 2, pp. 270–280, 1989.
[49] J. L. Elman, “Finding structure in time,” Cognit. Sci., Vol. 14, pp. 179–211, 1990.
[50] P. S. Sastry, G. Santharam, and K. P. Unnikrishnan, “Memory neural networks for identification and control of dynamic systems,” IEEE Trans. Neural Networks, vol. 5, pp. 306–319, Apr. 1994.
[51] Z. Michalewicz, Genetic Algorithms+Data Structures=Evolution Programs. New York: Springer-Verlag, 1999.
[52] J. Arabas, Z. Michalewicz, and J. Mulawka, “GAVaPS—A genetic algorithm with varying population size,” in Proc. IEEE Int. Conf. on Evolutionary Computation, Orlando, pp. 73–78, 1994.
[53] R. Tanese, “Distributed genetic algorithm,” in Proc. Int. Conf. Genetic Algorithms, pp. 434–439, 1989.
[54] R. J. Collins and D. R. Jefferson, “Selection in massively parallel genetic algorithms,” in Proc. Int. Conf. Genetic Algorithms, pp. 249–256, 1991.
[55] C. J. Lin,Y. J. Xu, “A novel evolution learning for recurrent wavelet-based neuro-fuzzy networks,” accepted to appear in Soft Computing, 2005.
[56] M. C. Hung and D. L. Yang, “The efficient fuzzy c-means clustering technique,” in Proc. IEEE Int. Conf. Data Mining, pp. 225 -232, Dec. 2001.
[57] J. He, L. Liu and G. Palm, “Speaker identification using hybrid LVQ-SLP networks,” in Proc. IEEE Int. Conf. Neural Networks, vol. 4, pp. 2052 -2055, Dec. 1995.
[58] K. K. Ang, C. Quek, and M. Pasquier, “POPFNN-CRI(S): Pseudo outer product based fuzzy neural network using the compositional rule of inference and singleton fuzzifier,” IEEE Trans. Syst., Man, Cybern., Part B, vol. 33, no. 6, pp. 838–849, Dec. 2003.
[59] A. G. Barto and R. S. Sutton, “Landmark learning: An illustration of associative search,” Biol. Cybern. Vol. 42, pp. 1-8, 1981.
[60] A. G. Barto, R. S. Sutton, and C. W. Anderson, “Neuronlike adaptive elements that can solve difficult learning control problem,” IEEE Trans. Syst., Man, Cybern., vol. SMC-13, no 5, pp. 834-847, 1983.
[61] K. C. Cheok and N. K. Loh, “A ball-balancing demonstration of optimal and disturbance-accommodating control,” IEEE Contr. Syst. Mag., pp. 54–57, 1987.
[62] R. H. Cannon, Jr., Dynamics of Physical Systems. New York: Mc- Graw-Hill, 1967.

[63] C. T. Lin and C. S. G. Lee, “Reinforcement structure/parameter learning for neural-network-based fuzzy logic control systems,” IEEE Trans. Fuzzy Syst., vol. 2, pp. 46–63, Feb. 1994.
[64] D. Whitley, S. Dominic, R. Das, and C. W. Anderson, “Genetic reinforcement learning for neuro control problems,” Mach. Learn., vol. 13, pp. 259–284, 1993.