### Foreword and Preface

• Foreword by Lotfi Zadeh xv
• Preface xix
• Audience xix
• Organization xx
• Features xxii
• Obtaining the Example Programs xxiii
• Acknowledgments xxiv

### Overview

• Chapter 1: Introduction to Neuro-Fuzzy and Soft Computing 1
• 1.1 Introduction 1
• 1.2 Soft Computing Constituents and Conventional Artificial Intelligence 1
• 1.3 Neuro-Fuzzy and Soft Computing Characteristics 7

### Part I: Fuzzy Set Theory

• Chapter 2: Fuzzy Sets 13
• 2.1 Introduction 13
• 2.2 Basic Definitions and Terminology 14
• 2.3 Set-theoretic Operations 21
• 2.4 MF Formulation and Parameterization 24
• 2.5 More on Fuzzy Union, Intersection, and Complement* 35
• 2.6 Summary 42
• Exercises 42
• Chapter 3: Fuzzy Rules and Fuzzy Reasoning 47
• 3.1 Introduction 47
• 3.2 Extension Principle and Fuzzy Relations 47
• 3.3 Fuzzy If-Then Rules 54
• 3.4 Fuzzy Reasoning 62
• 3.5 Summary 70
• Exercises 70
• Chapter 4: Fuzzy Inference Systems 73
• 4.1 Introduction 73
• 4.2 Mamdani Fuzzy Models 74
• 4.3 Sugeno Fuzzy Models 81
• 4.4 Tsukamoto Fuzzy Models 84
• 4.5 Other Considerations 85
• 4.6 Summary 89
• Exercises 90

### Part II: Regression and Optimization

• Chapter 5: Least-Squares Methods for System Identification 95
• 5.1 System Identification: An Introduction 95
• 5.2 Basics of Matrix Manipulation and Calculus 97
• 5.3 Least-Squares Estimator 104
• 5.4 Geometric Interpretation of LSE 110
• 5.5 Recursive Least-Squares Estimator 113
• 5.6 Recursive LSE for Time-Varying Systems* 116
• 5.7 Statistical Properties and the Maximum Likelihood Estimator* 118
• 5.8 LSE for Nonlinear Models 122
• 5.9 Summary 125
• Exercises 126
• Chapter 6: Derivative-based Optimization 129
• 6.1 Introduction 129
• 6.2 Descent Methods 129
• 6.3 The Method of Steepest Descent 133
• 6.4 Newton's Methods 134
• 6.5 Step Size Determination 141
• 6.6 Conjugate Gradient Methods* 148
• 6.7 Analysis of Quadratic Case 154
• 6.8 Nonlinear Least-squares Problems 160
• 6.9 Incorporation of Stochastic Mechanisms 166
• 6.10 Summary 168
• Exercises 168
• Chapter 7: Derivative-Free Optimization 173
• 7.1 Introduction 173
• 7.2 Genetic Algorithms 175
• 7.3 Simulated Annealing 181
• 7.4 Random Search 186
• 7.5 Downhill Simplex Search 189
• 7.6 Summary 193
• Exercises 194

### Part III: Neural Networks

• Chapter 8: Adaptive Networks 199
• 8.1 Introduction 199
• 8.2 Architecture 200
• 8.3 Backpropagation for Feedforward Networks 205
• 8.4 Extended Backpropagation for Recurrent Networks 210
• 8.5 Hybrid Learning Rule: Combining Steepest Descent and LSE 219
• 8.6 Summary 223
• Exercises 223
• Chapter 9: Supervised Learning Neural Networks 226
• 9.1 Introduction 226
• 9.2 Perceptrons 227
• 9.4 Backpropagation Multilayer Perceptrons 233
• 9.5 Radial Basis Function Networks 238
• 9.6 Modular Networks 246
• 9.7 Summary 250
• Exercises 251
• Chapter 10: Learning from Reinforcement 258
• 10.1 Introduction 258
• 10.2 Failure Is the Surest Path to Success 259
• 10.3 Temporal Difference Learning 264
• 10.4 The Art of Dynamic Programming 270
• 10.5 Adaptive Heuristic Critic 273
• 10.6 Q-learning 278
• 10.7 A Cost Path Problem 281
• 10.8 World Modeling* 288
• 10.9 Other Network Configurations* 290
• 10.10 Reinforcement Learning by Evolutionary Computation* 292
• 10.11 Summary 293
• Exercises 294
• Chapter 11: Unsupervised Learning and Other Neural Networks 301
• 11.1 Introduction 301
• 11.2 Competitive Learning Networks 302
• 11.3 Kohonen Self-Organizing Networks 305
• 11.4 Learning Vector Quantization 308
• 11.5 Hebbian Learning 310
• 11.6 Principal Component Networks 312
• 11.7 The Hopfield Network 316
• 11.8 Summary 327
• Exercises 328

### Part IV: Neuro-Fuzzy Modeling

• Chapter 12: ANFIS: Adaptive Neuro-Fuzzy Inference Systems 335
• 12.1 Introduction 335
• 12.2 ANFIS Architecture 336
• 12.3 Hybrid Learning Algorithm 340
• 12.4 Learning Methods that Cross-fertilize ANFIS and RBFN 341
• 12.5 ANFIS as a Universal Approximator* 342
• 12.6 Simulation Examples 345
• 12.7 Extensions and Advanced Topics 360
• Exercises 363
• Chapter 13: Coactive Neuro-Fuzzy Modeling: Towards Generalized ANFIS 369
• 13.1 Introduction 369
• 13.2 Framework 370
• 13.3 Neuron Functions for Adaptive Networks 372
• 13.4 Neuro-Fuzzy Spectrum 382
• 13.5 Analysis of Adaptive Learning Capability 385
• 13.6 Summary 393
• Exercises 395

### Part V: Advanced Neuro-Fuzzy Modeling

• Chapter 14: Classification and Regression Trees 403
• 14.1 Introduction 403
• 14.2 Decision Trees 404
• 14.3 CART Algorithm for Tree Induction 406
• 14.4 Using CART for Structure Identification in ANFIS 416
• 14.5 Summary 421
• Exercises 421
• Chapter 15: Data Clustering Algorithms 423
• 15.1 Introduction 423
• 15.2 K-Means Clustering 424
• 15.3 Fuzzy C-Means Clustering 425
• 15.4 Mountain Clustering Method 427
• 15.5 Subtractive Clustering 431
• 15.6 Summary 432
• Exercises 432
• Chapter 16: Rulebase Structure Identification 434
• 16.1 Introduction 434
• 16.2 Input Selection 435
• 16.3 Input Space Partitioning 436
• 16.4 Rulebase Organization 441
• 16.5 Focus Set--Based Rule Combination 446
• 16.6 Summary 447
• Exercises 448

### Part VI: Neuro-Fuzzy Control

• Chapter 17: Neuro-Fuzzy Control I 453
• 17.1 Introduction 453
• 17.2 Feedback Control Systems and Neuro-Fuzzy Control: An Overview 454
• 17.3 Expert Control: Mimicking An Expert 458
• 17.4 Inverse Learning 460
• 17.5 Specialized Learning 465
• 17.6 Backpropagation Through Time and Real-Time Recurrent Learning 469
• 17.7 Summary 476
• Exercises 477
• Chapter 18: Neuro-Fuzzy Control II 480
• 18.1 Introduction 480
• 18.2 Reinforcement Learning Control 480
• 18.4 Gain Scheduling 489
• 18.5 Feedback Linearization and Sliding Control 493
• 18.6 Summary 496
• Exercises 497

• Chapter 19: ANFIS Applications 503
• 19.1 Introduction 503
• 19.2 Printed Character Recognition 503
• 19.3 Inverse Kinematics Problems 507
• 19.4 Automobile MPG Prediction 510
• 19.5 Nonlinear System Identification 514
• 19.6 Channel Equalization 516
• 19.7 Adaptive Noise Cancellation 523
• Chapter 20: Fuzzy-Filtered Neural Networks 535
• 20.1 Introduction 535
• 20.2 Fuzzy-Filtered Neural Networks 536
• 20.3 Application 1: Plasma Spectrum Analysis 538
• 20.4 Application 2: Hand-Written Numeral Recognition 540
• 20.5 Genetic Algorithm--based Fuzzy Filters 543
• 20.6 Summary 549
• Chapter 21: Fuzzy Sets and Genetic Algorithms in Game Playing 551
• 21.1 Introduction 551
• 21.2 Variants of Genetic Algorithms 551
• 21.3 Using Genetic Algorithms in Game Playing 553
• 21.4 Simulation Results of the Basic Model 556
• 21.5 Using Fuzzily Characterized Features 559
• 21.6 Using Polyploid GA in Game Playing 560
• 21.7 Summary 564
• Chapter 22: Soft Computing for Color Recipe Prediction 568
• 22.1 Introduction 568
• 22.2 Color Recipe Prediction 569
• 22.3 Single MLP approaches 569
• 22.4 CANFIS modeling for Color Recipe Prediction 571
• 22.5 Color Paint Manufacturing Intelligence 577
• 22.6 Experimental Evaluation 587
• 22.7 Discussion 589
• 22.8 Concluding Remarks and Future Directions 591
• Appendix A: Hints to Selected Exercises 595
• Appendix B: List of Internet Resources 598
• Appendix C: List of MATLAB Programs 601
• Appendix D: List of Acronyms 604
• Index 607

The Book's Home , Roger's Home