Seiffertt Unified Computational Intelligence for Complex Systems

1. Auflage 2010
ISBN: 978-3-642-03180-9
Verlag: Springer
Format: PDF
Kopierschutz: Adobe DRM (»Systemvoraussetzungen)

E-Book, Englisch, 150 Seiten

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.

Autoren/Hrsg.

Seiffertt, John

Inhaltsverzeichnis

1;Contents;6
2;Introduction;9
2.1;The Need for Unified Computational Intelligence;9
2.2;Contributions of This Work;12
2.3;The Three Types of Machine Learning;12
2.3.1;Unsupervised Learning;12
2.3.2;Supervised Learning;16
2.3.3;Reinforcement Learning;17
2.3.3.1;Approximate Dynamic Programming;18
2.3.3.2;Markov Decision Processes;19
2.3.3.3;The Bellman Equation;20
2.3.3.4;Heuristic Dynamic Programming;21
2.4;A Unified Approach;22
2.5;Future Work;24
3;The Unified Art Architecture;26
3.1;Introduction;26
3.2;Motivation;26
3.3;Block Diagram;27
3.4;Operation;29
3.4.1;Step 1: Calculate State Trace;29
3.4.2;Step 2: Calculate Control;31
3.4.3;Step 3: Process Control;32
3.4.4;Step 4: Interpret Reward via Critic;33
3.4.4.1;Supervisory Signal;33
3.4.4.2;Positive Reinforcement;33
3.4.4.3;Negative Reinforcement;34
3.4.4.4;Unsupervised Mode;34
3.5;An Extended Architecture;34
3.5.1;The Vigilance Test;34
3.5.2;The Weight Update;37
3.5.3;Algorithm;39
4;An Application of Unified Computational Intelligence;40
4.1;Overview;40
4.2;Introduction;40
4.2.1;Machine Learning;41
4.2.2;Information Fusion;41
4.3;Approach;42
4.3.1;System Architecture;42
4.3.2;Information Fusion Engine;43
4.4;Application;44
4.4.1;Vehicle Tracking;47
4.4.2;Analysis;48
4.4.2.1;Force Protection Experiments;48
4.4.2.2;Results of Training the Fusion Model;50
4.5;Future Work;54
4.6;Conclusion;55
5;The Time Scales Calculus;56
5.1;Introduction;56
5.2;Fundamentals;57
5.3;Single-Variable Calculus;59
5.4;Calculus of Multiple Variables;62
5.5;Extension of the Chain Rule;63
5.6;Induction on Time Scales;65
5.7;Quantum Calculus;65
6;Approximate Dynamic Programming on Time Scales;68
6.1;Overview;68
6.2;Introduction;68
6.3;Dynamic Programming Overview;69
6.4;Dynamic Programming Algorithm on Time Scales;70
6.4.1;Delta Derivative Version;71
6.4.2;Quantum Calculus Version;73
6.5;HJB Equation on Time Scales;76
6.5.1;Delta Derivative Version;77
6.5.2;Nabla Derivative Version;79
6.5.3;Alpha Derivative Version;81
6.6;Conclusions;82
7;Backpropagation on Time Scales;84
7.1;Overview;84
7.2;Introduction;84
7.3;Ordered Derivatives;85
7.3.1;Network Definitions;86
7.3.2;Structure of Ordered Derivatives;87
7.3.3;The Chain Rule;89
7.4;The Backpropagation Algorithm on Time Scales;92
7.5;Quantum Calculus;93
7.6;Conclusions;96
8;Unified Computational Intelligence in Social Science;97
8.1;Introduction;97
8.2;Game Theory and Computational Social Science;98
8.2.1;Computational Intelligence;98
8.2.2;Agent-Based Computational Social Science;102
8.2.3;Game Theory;103
8.3;Economics and Finance;104
8.3.1;Introduction;104
8.3.2;Background;105
8.3.3;Agent-Based Computational Economics;105
8.3.4;Application to Economic Systems;107
8.3.5;Future Research Directions;108
8.4;Intelligence in Markets;108
8.4.1;Introduction;109
8.4.2;Approximate Dynamic Programming and Stochastic Control;110
8.4.3;Evolving Asset Pricing Strategies;112
8.4.4;The Design of Market Mechanisms;114
8.4.5;Computational Markets;115
9;References;116