Multi Agent System (MAS) is a subfield of Distributed Artificial Intelligence (DAI) that studies behaviors of groups of agents and the complexity of their interactions. The identification, design and implementation of strategies for coordination is a central research issue in the field of DAI. It is nearly impossible to identify or even prove the existence of the best coordination strategy. In most cases a coordination strategy is chosen if it is reasonably good. The task of hand-coding agent behaviors to achieve desired coordination and team behaviors is very difficult, if not intractable. On the other hand The complexity of multi agent problems can rise with the number of agents and their behavioral sophistication. The field of cooperative multi agent learning promises solutions to these issues by trying to discover agent behaviors and suggesting new approaches to these problems and as such it has been the focus of numerous studies in recent years. Genetic Network Programming (GNP) is a cooperative multi agent learning method that is proposed recently by iiration from Genetic Programming (GP). While GP uses tree structure for representation of solutions, GNP uses a network architecture which can improve solution representation and search ability.