Searching Good Strategies in Evolutionary Minority Game Using Variable Length Genetic Algorithm

WS Yang,BH Wang,YL Wu,YB Xie
DOI: https://doi.org/10.1016/j.physa.2004.03.065
2004-01-01
Abstract:We propose and study a new adaptation minority game for understanding the complex dynamical behavior characterized by agent interactions competing limited resource in many natural and social systems. We compare the strategy of agents in the model to chromosome in biology. In our model, the agents with poor performance during certain time period may modify their strategies via variable length genetic algorithm which consists of cut and splice operator, imitating similar processes in biology. The performances of the agents in our model are calculated for different parameter conditions and different evolution mechanism. It is found that the system may evolve into a much more ideal equilibrium state, which implies much stronger cooperation among agents and much more effective utilization of the social resources. It is also found that the distribution of the strategies held by agents will tend towards a state concentrating upon small m region.
What problem does this paper attempt to address?