Weak disorder asymptotics in the stochastic mean-field model of distance (Q2428045)

A stochastic model of weighted distances is considered. The underlying graph is the complete graph on \(n\) vertices and each pair \(e\) of distinct vertices is associated with a random variable \(E_e\), which is exponentially distributed with mean 1 and is independent of every other pair. The weight of the edge \(e\) is set to be equal to \((E_e)^s\), where \(s>0\) is a parameter of the system. The optimal path between two vertices is the path that minimizes the sum of weights among all possible paths between these vertices -- this sum is called the cost of the path. When \(s=1\), then the model reduces to the well-studied first-passage percolation process on the complete graph. But when \(s\rightarrow \infty\), then the weight of a path tends to the maximum of the weights among the edges that belong to the path -- this is called the strong disorder regime. The authors determine the asymptotic distribution of the cost of the optimal path between two given vertices as well as the distribution of the number of hops of such a path. The distribution of the cost of the optimal path is associated with a parameter of a continuous-time branching process with a certain offspring distribution.