Berlin 2012 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
SOE: Fachverband Physik sozio-ökonomischer Systeme
SOE 3: Economic Models and Evolutionary Game Theory I
SOE 3.5: Talk
Monday, March 26, 2012, 12:00–12:15, H 0110
Complex dynamics in game learning — •James Sanders1, Tobias Galla1, and J. Doyne Farmer2 — 1School of Physics and Astronomy, The University of Manchester, Manchester, UK — 2Santa Fe Institute, Santa Fe, New Mexico, USA
Game learning processes are of interest in many fields, from the study of financial markets in which traders learn from their past successes and failures, to the design of computer systems that can improve their behaviour based on previous experiences. We study experience-weighted attraction, an empirically-based model of game learning behaviour, focusing in particular on the conditions under which it displays chaotic dynamics. We discuss low-dimensional games as well as those with a large number of strategies, where very high-dimensional chaotic attractors can be seen. This has potential implications for complex systems that can be modelled using experience-weighted attraction, as the presence of this kind of complex dynamics places severe limits on the predictability of a system.