Unified Computational Intelligence for Complex Systems (eBook)

John Seiffertt, Donald C. Wunsch (Autoren)

eBook Download: PDF

2010
150 Seiten
Springer Berlin (Verlag)
978-3-642-03180-9 (ISBN)

Infos

Computational intelligence encompasses a wide variety of techniques that allow computation to learn, to adapt, and to seek. That is, they may be designed to learn information without explicit programming regarding the nature of the content to be retained, they may be imbued with the functionality to adapt to maintain their course within a complex and unpredictably changing environment, and they may help us seek out truths about our own dynamics and lives through their inclusion in complex system modeling. These capabilities place our ability to compute in a category apart from our ability to erect suspension bridges, although both are products of technological advancement and reflect an increased understanding of our world. In this book, we show how to unify aspects of learning and adaptation within the computational intelligence framework. While a number of algorithms exist that fall under the umbrella of computational intelligence, with new ones added every year, all of them focus on the capabilities of learning, adapting, and helping us seek. So, the term unified computational intelligence relates not to the individual algorithms but to the underlying goals driving them. This book focuses on the computational intelligence areas of neural networks and dynamic programming, showing how to unify aspects of these areas to create new, more powerful, computational intelligence architectures to apply to new problem domains.

Inhalt

Contents 6
Introduction 9
The Need for Unified Computational Intelligence 9
Contributions of This Work 12
The Three Types of Machine Learning 12
Unsupervised Learning 12
Supervised Learning 16
Reinforcement Learning 17
Approximate Dynamic Programming 18
Markov Decision Processes 19
The Bellman Equation 20
Heuristic Dynamic Programming 21
A Unified Approach 22
Future Work 24
The Unified Art Architecture 26
Introduction 26
Motivation 26
Block Diagram 27
Operation 29
Step 1: Calculate State Trace 29
Step 2: Calculate Control 31
Step 3: Process Control 32
Step 4: Interpret Reward via Critic 33
Supervisory Signal 33
Positive Reinforcement 33
Negative Reinforcement 34
Unsupervised Mode 34
An Extended Architecture 34
The Vigilance Test 34
The Weight Update 37
Algorithm 39
An Application of Unified Computational Intelligence 40
Overview 40
Introduction 40
Machine Learning 41
Information Fusion 41
Approach 42
System Architecture 42
Information Fusion Engine 43
Application 44
Vehicle Tracking 47
Analysis 48
Force Protection Experiments 48
Results of Training the Fusion Model 50
Future Work 54
Conclusion 55
The Time Scales Calculus 56
Introduction 56
Fundamentals 57
Single-Variable Calculus 59
Calculus of Multiple Variables 62
Extension of the Chain Rule 63
Induction on Time Scales 65
Quantum Calculus 65
Approximate Dynamic Programming on Time Scales 68
Overview 68
Introduction 68
Dynamic Programming Overview 69
Dynamic Programming Algorithm on Time Scales 70
Delta Derivative Version 71
Quantum Calculus Version 73
HJB Equation on Time Scales 76
Delta Derivative Version 77
Nabla Derivative Version 79
Alpha Derivative Version 81
Conclusions 82
Backpropagation on Time Scales 84
Overview 84
Introduction 84
Ordered Derivatives 85
Network Definitions 86
Structure of Ordered Derivatives 87
The Chain Rule 89
The Backpropagation Algorithm on Time Scales 92
Quantum Calculus 93
Conclusions 96
Unified Computational Intelligence in Social Science 97
Introduction 97
Game Theory and Computational Social Science 98
Computational Intelligence 98
Agent-Based Computational Social Science 102
Game Theory 103
Economics and Finance 104
Introduction 104
Background 105
Agent-Based Computational Economics 105
Application to Economic Systems 107
Future Research Directions 108
Intelligence in Markets 108
Introduction 109
Approximate Dynamic Programming and Stochastic Control 110
Evolving Asset Pricing Strategies 112
The Design of Market Mechanisms 114
Computational Markets 115
References 116

Produkt-Details

• Erscheint lt. Verlag: 15.7.2010
• Reihe/Serie: Adaptation, Learning, and Optimization
• Reihe/Serie: Adaptation, Learning, and Optimization
• Zusatzinfo: 150 p. 9 illus. in color.
• Verlagsort: Berlin
• Sprache: englisch
• Themenwelt: Mathematik / Informatik ► Informatik ► Netzwerke
• Themenwelt: Informatik ► Theorie / Studium ► Künstliche Intelligenz / Robotik
• Themenwelt: Naturwissenschaften ► Physik / Astronomie
• Themenwelt: Technik
• Themenwelt: Wirtschaft
• Schlagworte: Adaptive Resonance Theory • Agent-Based Computational Social Science • Approximate Dynamic Programming • backpropagation • Backpropogation • Clustering • Complexity • Computational Intelligence • Dynamic Equations • Game Theory • learning • Modeling • neural network • Neural networks • Reinforcement Learning • supervised learning • Time Scales Calculus • Unsupervised Learning
• ISBN-10: 3-642-03180-3 / 3642031803
• ISBN-13: 978-3-642-03180-9 / 9783642031809