Advanced Fuzzy Systems Design and ApplicationsSpringer Science & Business Media, 2003 - 271 páginas Fuzzy rule systems have found a wide range of applications in many fields of science and technology. Traditionally, fuzzy rules are generated from human expert knowledge or human heuristics for relatively simple systems. In the last few years, data-driven fuzzy rule generation has been very active. Compared to heuristic fuzzy rules, fuzzy rules generated from data are able to extract more profound knowledge for more complex systems. This book presents a number of approaches to the generation of fuzzy rules from data, ranging from the direct fuzzy inference based to neural net works and evolutionary algorithms based fuzzy rule generation. Besides the approximation accuracy, special attention has been paid to the interpretabil ity of the extracted fuzzy rules. In other words, the fuzzy rules generated from data are supposed to be as comprehensible to human beings as those generated from human heuristics. To this end, many aspects of interpretabil ity of fuzzy systems have been discussed, which must be taken into account in the data-driven fuzzy rule generation. In this way, fuzzy rules generated from data are intelligible to human users and therefore, knowledge about unknown systems can be extracted. |
Índice
Fuzzy Sets and Fuzzy Systems | 1 |
112 Fuzzy Operations | 7 |
113 Fuzzy Relations | 10 |
114 Measures of Fuzziness | 13 |
115 Measures of Fuzzy Similarity | 15 |
12 Fuzzy Rule Systems | 16 |
122 Fuzzy Rules for Modeling and Control | 19 |
123 Mamdani Fuzzy Rule Systems | 25 |
524 Interpretability Issues | 120 |
53 Modeling and Control Using the Neurofuzzy System | 123 |
532 Dynamic Robot Control | 124 |
54 Neurofuzzy Control of Nonlinear Systems | 130 |
541 Fuzzy Linearization | 132 |
542 Neurofuzzy Identification of the Subsystems | 135 |
543 Design of Controller | 137 |
544 Stability Analysis | 138 |
124 TakagiSugenoKang Fuzzy Rule Systems | 26 |
125 Fuzzy Systems are Universal Approximators | 27 |
13 Interpretability of Fuzzy Rule System | 29 |
132 The Properties of Membership Functions | 30 |
134 Distinguishability of Fuzzy Partitions | 33 |
135 Consistency of Fuzzy Rules | 34 |
136 Completeness and Compactness of Rule Structure | 37 |
142 Knowledge Representation with Fuzzy Preference Models | 42 |
143 Fuzzy Group Decision Making | 45 |
Evolutionary Algorithms | 49 |
221 Representation | 50 |
222 Recombination | 53 |
223 Mutation | 54 |
224 Selection | 55 |
232 Selfadaptation | 56 |
24 Constraints Handling | 58 |
25 Multiobjective Evolution | 60 |
251 Weighted Aggregation Approaches | 61 |
252 Populationbased NonPareto Approaches | 62 |
254 Discussions | 63 |
26 Evolution with Uncertain Fitness Functions | 64 |
263 Robustness Considerations | 68 |
27 Parallel Implementations | 69 |
28 Summary | 70 |
Artificial Neural Networks | 73 |
321 Multilayer Perceptrons | 74 |
322 Radial Basis Function Networks | 75 |
331 Supervised Learning | 76 |
332 Unsupervised Learning | 78 |
333 Reinforcement Learning | 79 |
34 Improvement of Generalization | 80 |
341 Heuristic Methods | 81 |
343 Regularization | 82 |
344 Network Ensembles | 84 |
345 A Priori Knowledge | 85 |
35 Rule Extraction from Neural Networks | 86 |
352 Extraction of Fuzzy Rules | 87 |
36 Interaction between Evolution and Learning | 89 |
37 Summary | 90 |
Conventional Datadriven Fuzzy Systems Design | 93 |
42 Fuzzy Inference Based Method | 94 |
43 WangMendels Method | 100 |
44 A Direct Method | 102 |
45 An Adaptive Fuzzy Optimal Controller | 105 |
46 Summary | 110 |
Neural Network Based Fuzzy Systems Design | 111 |
52 The Pisigma Neurofuzzy Model | 114 |
522 The Hybrid Neural Network Model | 115 |
523 Training Algorithms | 116 |
55 Summary | 141 |
Evolutionary Design of Fuzzy Systems | 143 |
62 Evolutionary Design of Flexible Structured Fuzzy Controller | 145 |
622 Parameter Optimization Using Evolution Strategies | 146 |
623 Simulation Study | 147 |
63 Evolutionary Optimization of Fuzzy Rules | 148 |
632 Fitness Function | 152 |
633 Evolutionary Fuzzy Modeling of Robot Dynamics | 153 |
64 Fuzzy Systems Design for HighDimensional Systems | 160 |
642 Flexible Fuzzy Partitions | 161 |
643 Hierarchical Structures | 163 |
644 Input Dimension Reduction | 164 |
645 GABased Input Selection | 169 |
65 Summary | 171 |
Knowledge Discovery by Extracting Interpretable Fuzzy Rules | 173 |
712 Interpretability of Fuzzy Systems and Knowledge Extraction | 174 |
72 Evolutionary Interpretable Fuzzy Rule Generation | 175 |
721 Evolution Strategy for Mixed Parameter Optimization | 176 |
722 Genetic Representation of Fuzzy Systems | 177 |
723 Multiobjective Fuzzy Systems Optimization | 178 |
Fuzzy Vehicle Distance Controller | 180 |
73 Interactive Coevolution for Fuzzy Rule Extraction | 184 |
732 Coevolution | 186 |
74 Fuzzy Rule Extraction from RBF Networks | 187 |
742 Fuzzy Rule Extraction by Regularization | 191 |
743 Application Examples | 196 |
75 Summary | 203 |
Fuzzy Knowledge Incorporation into Neural Networks | 205 |
82 Knowledge Incorporation in Neural Networks for Control | 207 |
822 Knowledge Incorporation in Adaptive Neural Control | 208 |
83 Fuzzy Knowledge Incorporation By Regularization | 210 |
84 Fuzzy Knowledge as A Related Task in Learning | 216 |
85 Simulation Studies | 217 |
851 Regularized Learning | 218 |
852 Multitask Learning | 219 |
86 Summary | 221 |
Fuzzy Preferences Incorporation into Multiobjective Optimization | 223 |
912 Incorporation of Fuzzy Preferences | 225 |
92 Evolutionary Dynamic Weighted Aggregation | 226 |
921 Conventional Weighted Aggregation for MOO | 227 |
922 Dynamically Weighted Aggregation | 228 |
923 Archiving of Pareto Solutions | 230 |
925 Theoretical Analysis | 237 |
93 Fuzzy Preferences Incorporation in MOO | 247 |
932 Converting Fuzzy Preferences into Weight Intervals | 249 |
94 Summary | 252 |
255 | |
269 | |
Outras edições - Ver tudo
Palavras e frases frequentes
adaptation approach approximate model basis functions chromosome coding concave Pareto front constraints convex data pairs defined defuzzification denoted distance distribution dynamics equation error evolution control evolution strategies evolutionary algorithms evolutionary computation example fitness function fitness space fuzzy controller fuzzy inference fuzzy membership functions fuzzy partition fuzzy relation fuzzy rule system fuzzy set fuzzy subsets fuzzy system Gaussian Gaussian function genetic algorithms heuristics hidden nodes IEEE Transactions input variables interpretable fuzzy rules j-th learning algorithm linguistic terms linguistic variable Manhattan distance multiobjective optimization mutation mutual information neurofuzzy model neurofuzzy systems neuron nonlinear system number of fuzzy obtained output parameter space Pareto front Pareto optimal Pareto solutions performance population priori knowledge problem RBF network recombination related task robot rule base rule extraction rule structure samples sensors shown in Fig simulation sub-population supervised learning T-norm test function tion training data TSK fuzzy rules vector weighted aggregation method
Passagens conhecidas
Página 255 - When both individuals and populations search: Adding simple learning to the genetic algorithm.
Página 255 - JM Benitez, JL Castro, and I. Requena, "Are Artificial Neural Networks Black Boxes?", IEEE Transactions on Neural Networks, Vol.
Referências a este livro
Computational Intelligence, Theory and Applications: International ... Bernd Reusch Pré-visualização limitada - 2006 |