Title: | Collectives and the Design of Complex Systems |

Editor: | Kagan Tumer and David Wolpert |

Publisher: | SPringer |

Year: | 2004 |

ISBN: | 0387401652 |

Abstract: | With the advent of extremely affordable computing power, the world is becoming filled with distributed systems of computationally sophisticated components. However, no current scientific discipline offers a thorough understanding of the relation of such "collectives" and how well they meet performance criteria. Collectives and Design of Complex Systems lays the foundation for the study of collective intelligence and how these entities can be developed to yield optimal performance. Part one describes how some information-processing problems can only be solved by the joint actions of large communities of computers, each running their own complex, decentralized machine-learning algorithms. Part two offers general analysis on the dynamics and structures of collectives. Finally, part three addresses economic, model-free, and control-theory approaches to designing these complex systems. The work assumes a modest understanding of basic statistics and calculus. Integrates theory with real-world practice. |

Cited by 12

@Book{tumer04a, editor = {Kagan Tumer and David Wolpert}, title = {Collectives and the Design of Complex Systems}, publisher = {SPringer}, year = 2004, abstract = {With the advent of extremely affordable computing power, the world is becoming filled with distributed systems of computationally sophisticated components. However, no current scientific discipline offers a thorough understanding of the relation of such "collectives" and how well they meet performance criteria. Collectives and Design of Complex Systems lays the foundation for the study of collective intelligence and how these entities can be developed to yield optimal performance. Part one describes how some information-processing problems can only be solved by the joint actions of large communities of computers, each running their own complex, decentralized machine-learning algorithms. Part two offers general analysis on the dynamics and structures of collectives. Finally, part three addresses economic, model-free, and control-theory approaches to designing these complex systems. The work assumes a modest understanding of basic statistics and calculus. Integrates theory with real-world practice.}, keywords = {multiagent learning complexity}, isbn = {0387401652}, googleid = {cx0h0fzYb7YJ:scholar.google.com/}, googleprint = {O-Xw23-eOWAC}, cluster = {13145964417669733747} }Last modified: Wed Mar 9 10:16:17 EST 2011