Using Distributed Learning Automata to Find Minimum Spanning Tree in Stochastic Graph

In this paper an algorithm based on distributed learning automata for finding minimum spanning tree in a stochastic graph when the weight density functions for the edges of the graph is unknown is propped. The proposed algorithm with the aid of distributed learning automata tries to find the minimum spanning tree with minimum number of sampling from the edges of the stochastic graph. Using simulation it has been shown that with proper selection of parameters of the distributed learning automata, the propped algorithm is able to find the minimum spanning tree with high propobability. As the algorithm proceeds, the process of sampling from the graph is concentrated on the edges of the minimum spanning tree with minimum expected cost. To evaluate the proposed algorithm, the number of sampling taken by the proposed algorithm is compared with the number of sampling needed by the standard sampling method. The result of comparison shows the efficiency of the proposed algorithm in terms of the number of samplings.