Second Parrondo's Paradox in Scale Free Networks (1207.4911v2)

Abstract: Parrondo's paradox occurs in sequences of games in which a winning expectation value of a payoff may be obtained by playing two games in a random order, even though each game in the sequence may be lost when played individually.Several variations of Parrondo's games apparently with the same paradoxical property have been introduced by G.P. Harmer and D. Abbott; history dependence, one dimensional line, two dimensional lattice and so on. I have shown that Parrondo's paradox does not occur in scale free networks in the simplest case with the same number of parameters as the original Parrondo's paradox. It suggests that some technical complexities are needed to present Parrondo's paradox in scale free networks. In this article, I show that a simple modification with the same number of parameters as the original Parrondo's paradox creates Parrondo's paradox in scale free. This paradox is, however, created by a quite different mechanism from the original Parrondo's paradox and a considerably rare phenomenon, where the discrete property of degree of nodes is crucial. I call it the second Parrondo's paradox.