Regensburg 2004 – scientific programme

Parts | Days | Selection | Search | Downloads | Help

AKSOE: Physik sozio-ökonomischer Systeme

AKSOE 3: Postersitzung

AKSOE 3.6: Poster

Monday, March 8, 2004, 16:00–18:00, Poster D

Investors’ Game Strategies using Genetic Algorithms — •J. Emeterio Navarro B.¹ and Frank Schweitzer^2,3 — ¹Institute for Informatics, Humboldt University, Berlin — ²Institute for Physics, Humboldt University, Berlin — ³Fraunhofer Institute for Autonomous Intelligent Systems, Schloss Birlinghoven, Sankt Augustin

We use an Investors’ Game model to study two kind of strategies investors can use: the first one is based on feedback information or previous experience and the second one is based on genetic algorithms. We compare these two approaches within a multi-agent system where agents play the role of investors and they interact with each other in order to establish projects. We investigate (i) the competition between these two different strategies, (ii) the formation of coalitions and (iii) how the learning process of agents depends on the genetic algorithm parameters.