臺灣博碩士論文加值系統

本研究針對模糊類神經網路來發展一套適應性學習演算法，用以解決類神經網路學習過程中的兩個重要議題：如何減少模糊類神經網路過適化現象，與其所造成準確率下降的問題。本研究提出配置證據架構的方法來減少模糊類神經網路過適化現象，加上其擁有倒傳遞學習機制，所以能夠有效提高模糊類神經網路的性能，即在學習過程的每一次迭代中，能夠權衡最小化期望錯誤值或模式複雜度來決定出適當的學習權重。最後透過一些模擬結果來展示本研究所建立的方法是一個有效的方法。

An adaptive learning algorithm for adapting fuzzy neural networks is developed in the study to deal with two important issues in the machine learning process: how to reduce the overfitting effect and how to improve the accuracy of prediction of fuzzy neural networks. This approach is able to reduce the overfitting effect with evidence framework, and to enhance fuzzy neural networks with backpropagation learning mechanism, i.e., determining the learning weight which is a trade-off by minimizing the mean error or decreasing the model complexity at each iteration of learning process. Finally, several simulations are conducted to demonstrate the efficiency of the proposed algorithm.

CHINESE ABSTRACT I
ABSTRACT II
ACKNOWLEDGEMENTS III
TABLE OF CONTENTS IV
LIST OF FIGURES VI
LIST OF TABLES VII
CHAPTER 1 INTRODUCTION 1
1.1 RESEARCH BACKGROUND 1
1.2 RESEARCH MOTIVATION 3
1.3 RESEARCH OBJECTIVES 3
1.4 RESEARCH LIMITATIONS 4
CHAPTER 2 LITERATURE REVIEW 5
2.1 THE BACKPROPAGATION ALGORITHM 5
2.2 THE EVIDENCE FRAMEWORK 6
2.2.1 Regression Problem 6
2.2.2 Classification Problem 10
2.3 THE FNN ARCHITECTURE 11
2.3.1 Fuzzy Logic System 11
2.3.2 The FALCON Model 14
2.3.3 Generating Fuzzy Rules From Numerical Data 21
CHAPTER 3 RESEARCH APPROACH 23
3.1 THE RESEARCH ARCHITECTURE 23
3.2 MACKEY-GLASS TIME SERIES PREDICTION 24
3.3 AN FNN DESIGNED FOR PREDICTION PROBLEMS 25
3.3.1 The FNN Architecture 25
3.3.2 The Structure Learning Phase for FNN 28
3.3.3 The Parameter Learning Phase for FNN 28
3.3.4 Model Fitness Adaption 31
CHAPTER 4 SIMULATIONS 32
4.1 AN NN WITHOUT THE EVIDENCE FRAMEWORK 32
4.2 AN NN WITH THE EVIDENCE FRAMEWORK 33
4.3 AN FNN WITHOUT THE EVIDENCE FRAMEWORK 34
4.4 AN FNN WITH THE EVIDENCE FRAMEWORK 34
4.5 DISCUSSION 35
CHAPTER 5 CONCLUSIONS 36
References 37

Altug, S., Trussell, H. J. & Chow, M. -Y. (1999). A “mutual update” training algorithm for fuzzy adaptive logic control/decision network (FALCON), IEEE Transactions on Neural Networks, 10(1), 196-199.
Archer, N. P. & Wang, S. (1991). Fuzzy set representation of neural network classification boundaries, IEEE Transactions on Systems, Man and Cybernetics, 21(4), 735-742.
Bakirtzis, A. G., Theocharis, J. B., Kiartzis, S. J. & Satsios, K. J. (1995). Short term load forecasting using fuzzy neural networks, IEEE Transactions on Power Systems, 10(3), 1518-1524.
Chung, F. -L. & Duan, J. -C. (2000). On multistage fuzzy neural network modeling, IEEE Transactions on Fuzzy Systems, 8(2), 125-142.
Colace, L., Giona, M. & Coli, M. (1995). Parallel implementation of a general purpose functional predictor for chaotic time series, Signal Processing, 41, 1-15.
Dash, P. K., Ramakrishna, G., Liew, A. C. & Rahman, S. (1995). Fuzzy neural networks for time-series forecasting of electric load, IEE Proceedings－Generation, Transmission and Distribution, 142(5), 535-544.
Foresee, F. D. & Hagan, M. T. (1997). Gauss-Newton approximation to Bayesian learning, Proceedings of the IEEE International Conference on Neural Networks at Houston, 3, 1930-1935.
Forster, M. R. (2000). Key concepts in model selection: performance and generalizability, Journal of Mathematical Psychology, 44, 205-231.
Gestel, T. V., Suykens, J. A. K., Baestaens, D. -E., Lambrechts, A., Lanckriet, G., Vandaele B., Moor, B. D. & Vandewalle, J. (2001). Financial time series prediction using least squares support vector machines within the evidence framework, IEEE Transactions on Neural Networks, 12(4), 809-821.
Hagan, M. T., Demuth, H. B. & Beale, M. (1996). Neural Network Design, Boston: PWS Publishing Company.
Kosko, B. (1990). Unsupervised learning in noise, IEEE Transactions on Neural Networks, 1(1), 44-57.
Kwok, J. T. -Y. (1999). Moderating the outputs of support vector machine classifiers, IEEE Transactions on Neural Networks, 10(5), 1018-1031.
Kwok, J. T. -Y. (2000). The evidence framework applied to support vector machines, IEEE Transactions on Neural Networks, 11(5), 1162-1173.
Lee, C. C. (1990a). Fuzzy logic in control systems: fuzzy logic controller－part I, IEEE Transactions on Systems, Man and Cybernetics, 20(2), 404-418.
Lee, C. C. (1990b). Fuzzy logic in control systems: fuzzy logic controller－part II, IEEE Transactions on Systems, Man and Cybernetics, 20(2), 419-435.
Lin, C. -J. & Lin, C. -T. (1996). Reinforcement learning for an ART-based fuzzy adaptive learning control network, IEEE Transactions on Neural Networks, 7(3), 709-731.
Lin, C. -J. & Lin, C. -T. (1997). An ART-based fuzzy adaptive learning control network, IEEE Transactions on Fuzzy Systems, 5(4), 477-496.
Lin, C. -T. & Lee, C. S. G. (1991). Neural-network-based fuzzy logic control and decision system, IEEE Transactions on Computers, 40(12), 1320-1336.
Lin, Y. & Cunningham III, G. A. (1995). A new approach to fuzzy-neural system modeling, IEEE Transactions on Fuzzy Systems, 3(2), 190-198.
Liu, P. & Li, H. (2004). Efficient learning algorithms for three-layer regular feedforward fuzzy neural networks, IEEE Transactions on Neural Networks, 15(3), 545-558.
MacKay, D. J. C. (1992a). Bayesian interpolation, Neural Computation, 4(3), 415-447.
MacKay, D. J. C. (1992b). A practical Bayesian framework for backpropagation networks, Neural Computation, 4(3), 448-472.
MacKay, D. J. C. (1992c). Information-based objective functions for active data selection, Neural Computation, 4(4), 590-604.
MacKay, D. J. C. (1992d). The evidence framework applied to classification networks, Neural Computation, 4(5), 698-714.
MacKay, D. J. C. (1995). Probable networks and plausible predictions－a review of practical Bayesian methods for supervised neural networks, Network: Computation in Neural Systems, 6(3), 469-505.
MacKay, D. J. C. (2003). Information Theory, Inference, and Learning Algorithms, Cambridge: Cambridge University Press.
Mackey, M. C. & Glass, L. (1977). Oscillation and chaos in physiological control systems, Science, 197, 287-289.
Maguire, L. P., Roche, B., McGinnity, T. M. & McDaid, L. J. (1998). Predicting a chaotic time series using a fuzzy neural network, Information Sciences, 112, 125-136.
Riedmiller, M. & Braun, H. (1993). A direct adaptive method for faster backpropagation learning: the RPROP algorithm, Proceedings of the IEEE International Conference on Neural Networks at San Francisco, 1, 586-591.
Shen, V. R. L. (2003). Reinforcement learning for high-level fuzzy petri nets, IEEE Transactions on Systems, Man and Cybernetics-Part B: Cybernetics, 33(3), 351-362.
Tung, W. L. & Quek, C. (2004). Falcon: neural fuzzy control and decision systems using FKP and PFKP clustering algorithms, IEEE Transactions on Systems, Man and Cybernetics-Part B: Cybernetics, 34(1), 686-695.
Vehtari, A. & Lampinen, J. (2000). Bayesian MLP neural networks for image analysis, Pattern Recognition Letters, 21, 1183-1191.
Wang, L. -X. & Mendel, J. M. (1992). Generating fuzzy rules by learning from examples, IEEE Transactions on Systems, Man and Cybernetics, 22(6), 1414-1427.
Wright, W. A. (1999). Bayesian approach to neural-network modeling with input uncertainty, IEEE Transactions on Neural Networks, 10(6), 1261-1270.
Zadeh, L. A. (1965). Fuzzy sets, Information Control, 8, 338-353.
Zadeh, L. A. (1975a). The concept of a linguistic variable and its application to approximate reasoning I, Information Science, 8, 199-251.
Zadeh, L. A. (1975b). The concept of a linguistic variable and its application to approximate reasoning II, Information Science, 8, 301-357.
Zadeh, L. A. (1975c). The concept of a linguistic variable and its application to approximate reasoning III, Information Science, 9, 43-80.
Zhang, J. & Morris, A. J. (1995). Fuzzy neural networks for nonlinear systems modeling, IEE Proceedings－Control Theory and Applications, 142(6), 551-561.