A central issue in the design of cooperative multiagent systems is how to c
oordinate the behavior of the agents to meet the goals of the designer. Tra
ditionally, this had been accomplished by hand-coding the coordination stra
tegies. However, this task is complex due to the interactions that can take
place among agents. Recent work in the area has focused on how strategies
can be learned. Yet, many of these systems suffer from convergence, complex
ity and performance problems. This paper presents a new approach for learni
ng multiagent coordination strategies that addresses these issues. The effe
ctiveness of the technique is demonstrated using a synthetic domain and the
predator and prey pursuit problem.