Strategy updating rules and strategy distributions in dynamical multiagent systems

Abstract

In the evolutionary version of the minority game, agents update their strategies (gene-value $p$) in order to improve their performance. Motivated by recent intriguing results obtained for prize-to-fine ratios which are smaller than unity, we explore the system's dynamics with a strategy updating rule of the form $p \to p \pm \delta p$ ($0 \leq p \leq 1$). We find that the strategy distribution depends strongly on the values of the prize-to-fine ratio $R$, the length scale $\delta p$, and the type of boundary condition used. We show that these parameters determine the amplitude and frequency of the the temporal oscillations observed in the gene space. These regular oscillations are shown to be the main factor which determines the strategy distribution of the population. In addition, we find that agents characterized by $p={1 \over 2}$ (a coin-tossing strategy) have the best chances of survival at asymptotically long times, regardless of the value of $\delta p$ and the boundary conditions used.