臺灣博碩士論文加值系統

本論文主要在研究實數編碼排序式基因法則，其染色體使用浮點數形式參數，此方法的排列式適應函數可以增加族群的變異性，此外在演化過程中，採用排序式重生及實數型交配、突變來產生子代。本論文還提出一種結合實數編碼排列式基因法則和倒傳遞演算法的新方法-演化型類神經網路，此類神經網路將人類的特性學習行為融合到演化當中。它不僅可改善倒傳遞演算法容易掉入區域最小解的缺點而且能夠克服基因法則無法有效收斂至鄰近區域最小解的困難。最後，將實數編碼排列式基因法則應用在尋找狀態回授控制器參數的問題上以展現它搜尋能力，也將演化型類神經網路應用在有名的或斥問題上來顯示它的優點。

This thesis presents a Real-Coded Rank-Based Genetic Algorithm (RCRBGA), which is represented by a chromosome containing parameters in floating-point. The use of rank-based fitness increases the population diversity. The offspring are generated by the rank-based reproduction, real parametric crossover and mutation in the evolving process. Besides, an Evolutionary Neural Network (ENN) which combines RCRBGA and Back-Propagation (BP) is introduced. ENN applies the learning concept to the evolution process, like the behavior of human beings. It not only improves the disadvantage of easily slumping in to local minima of BP but also overcomes the defect of genetic algorithm, which can't efficiently converge to minima. Finally, the search ability of RCRBGA is demonstrated by an example, linear state-feedback controller via pole-assignment method. In addition, ENN applies to a classifying problem of the modified XOR to show its advantage.

Chinese Abstract i
English Abstract ii
Acknowledgment iii
Contents iv
Index of Figures vii
Index of Tables viii
Notations ix
Chapter 1 Introduction
.. 1
1.1 Research motivation and purpose………………………………………1
1.2 Organization of the thesis………………………………………………4
Chapter 2 Basic Concepts of Genetic Algorithm and Neural Network.. 5
2.1 History of GA and NN………………………………………………….. 5
2.2 Biological Terminology and Basic Operators of GA…………………8
2.3 The Conventional Genetic Algorithm…………………………………..11
2.4 Basic Elements of Neural Network……………………………………..15
2.5 Back-Propagation……………………………………………………….. 18
Chapter 3 Real-Code Rank-Based Genetic Algorithm
.. 26
3.1 Specialized operators and concepts……………………………………26
3.1.1 Expression of Population……..…………………………….26
3.1.2 Reinitialization Process…………………………………….27
3.1.3 Rank-based Fitness………………..………………………..27
3.1.4 Rank-based Reproduction…………………………………... 28
3.1.5 Real Parametric Crossover and Mutation………………….29
3.1.6 Age and Lifetime………………………………………….... 30
3.2 Real-Code Rank-Based Genetic Algorithm.………….……………….. 31
3.3 Flowchart…………..………………………………………………….. 34
Chapter 4 Evolutionary Neural Network
.. 35
4.1 Specialized operators and concepts…………………………………..35
4.1.1 Search-Converge Criterion…………………….…………..35
4.1.2 Pocket Algorithm…………………………….…………….. 37
4.2 Evolutionary Neural Network………………………….……………... 37
4.3 Flowchart………………………………………….…………………... 40
Chapter 5 Applications
.. 41
5.1 Pole-Assignment Method……………………………………………... 41
5.1.1 Problem statement………………………………………….. 41
5.1.2 RCRBGA Implementation……………………….……..….. 42
5.1.3 Simulation Results………………………………….……...44
5.1.4 Analysis…………………………………………………….. 48
5.2 ENN used as the pattern classifier……………………………….…..49
5.2.1 Problem Statement………………………………………..... 49
5.2.2 ENN Implementation……………………………………..... 50
5.2.3 Simulation Results…………………………………….…...51
5.2.4 Analysis…………………………………………………...... 54
Chapter 6 Conclusions
.. 55
References57

[1] C. G. Looney, Pattern Recognition Using Neural Networks, NY, USA, Oxford University Press, 1997.
[2] S. Haykin, Neural Networks, NY, USA, Macmillan College Press,1994.
[3] D. E. Goldberg, Genetic algorithms in Search, Optimization and Machine learning, Addison-Wesley, 1989.
[4] M. Mitchell, An Introduction to Genetic Algorithm, USA, MIT Press, 1997.
[5] Z. Michalewicz, Genetic Algorithms + Data Structures = Evolution programs, Springer-Verlag, 1994.
[6] K. F. Man, K. S. Tang and S. Kwong, "Genetic Algorithms: Concepts and Applications," IEEE Transactions on Industrial Electronics, Vol.5, pp.519-534, 1996.
[7] Y. Li, K. C. Ng and D. J. Murray-Smith, "Genetic algorithm automated approach to the design of sliding mode control systems," Journal of Control, Vol.4, pp. 721-739, 1996.
[8] J. J. Grefenstette, "Optimization of Control Parameters for Genetic Algorithms", IEEE Transactions on Systems, Man, and Cybernetics, Vol.16, No. 1, pp.122-128, 1986.
[9] D. Whitley, "The GENITOR Algorithm and Selection Pressure: Why Rank-based Allocation of Reproductive Trials is Best," The Third International Conference on Genetic Algorithms, Los Atlos, CA, pp.116-121, 1989.
[10] Y. P. Chen and K.T. Li, "Real-Coded Rank-Based Genetic Algorithm," 自動控制研討會, pp. 162-164, 1998.
[11] M. Gori and A. Tesi, "On the Problem of Local Minima in Backpropagation," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 14, No.1, pp. 76-85, 1992.
[12] A. D. Brown and H. C. Card, "Evolutionary Artificial Neural Networks for Competitive Learning," International Conference on Neural Networks, pp.1558-1562, 1997.
[13] S. H. Leung, A. Luk and S. C. Ng, "Fast convergent genetic-type search for multi-layered network," IEICE Trans. On Fundamental of Electronics, Communications and computer sciences, Vol. E77-A, No.9, pp.1484-1492, 1994.
[14] S. C. Ng, S. H. Leung and A. Luk, "Fast and global convergent weight evolution algorithm based on modified back-propagation," Proc. Of IEEE International Conference on Neural Networks, pp.3004-3008, 1995.
[15] S. C. Ng, S. H. Leung and A. Luk, "Weight Evolution Algorithm with Dynamic Offset Range," Proc. Of IEEE International Conference on Neural Networks, pp. 1093-1096, 1997.
[16] Y. P. Chen and K.T. Li, "A new method combined Genetic Algorithm and Neural Network," 自動控制研討會, pp. 162-164, 1999.
[17] J. Abras, Z. Michalewicz and J. Mulawka, "GAVaPS- a Genetic Algorithm with Varying Population Size," Proceedings of the Evolutionary Computation Conference, Orlando, pp. 73-78, 1994.
[18] C. Darken and J. Moody, "Towards faster stochastic gradient search," In Advances in Neural Information Processing Systems, pp. 1009-1016, San Mateo, CA:Morgan Kaufmann, 1992.
[19] S. I. Gallant, Neural Network Learning and Expert Systems, The MIT Press, Massachusetts, 1993.
[20] E. K. Blum, "Approximation of Boolean functions by sigmoidal networks: Part Ⅰ: XOR and other two-variable functions," Neural computation, Vol. 1, pp.532-540, 1989.