Home // EMERGING 2010, The Second International Conference on Emerging Network Intelligence // View article

A Nontraditional Approach for a Highly Interactive Collective-Adaptive System

Authors:
Dávid Lányi
Borbála Katalin Benko

Keywords: collective-adaptive systems, open learning

Abstract:
In this paper, we describe a nontraditional approach for realizing a highly interactive adaptive system with a mini-mal level of intrinsic knowledge; which opens up the way for a more open and active adaptation to the rules and dynamic changes of the environment. First, we discuss a lightweight, incremental adaptation model where knowledge and strategy emerge during (and throughout) the system’s normal opera-tion. Then, this model is extended with a collective mechanism: a society of learners makes use of each other’s findings in order to converge faster. Finally, principles are evaluated with simulation in a theoretical showcase (‘Connect-5’) world.

Pages: 101 to 106

Copyright: Copyright (c) IARIA, 2010

Publication date: October 25, 2010

Published in: conference

ISSN: 2326-9383

ISBN: 978-1-61208-103-8

Location: Florence, Italy

Dates: from October 25, 2010 to October 30, 2010