Evolving Team Compositions by Agent Swapping

Optimizing collective behavior in multiagent systems requires algorithms to find not only appropriate individual behaviors but also a suitable composition of agents within a team. Over the last two decades, evolutionary methods have emerged as a promising approach for the design of agents and their compositions into teams. The choice of a crossover operator that facilitates the evolution of optimal team composition is recognized to be crucial, but so far, it has never been thoroughly quantified. Here, we highlight the limitations of two different crossover operators that exchange entire agents between teams: restricted agent swapping (RAS) that exchanges only corresponding agents between teams and free agent swapping (FAS) that allows an arbitrary exchange of agents. Our results show that RAS suffers from premature convergence, whereas FAS entails insufficient convergence. Consequently, in both cases, the exploration and exploitation aspects of the evolutionary algorithm are not well balanced resulting in the evolution of suboptimal team compositions. To overcome this problem, we propose combining the two methods. Our approach first applies FAS to explore the search space and then RAS to exploit it. This mixed approach is a much more efficient strategy for the evolution of team compositions compared to either strategy on its own. Our results suggest that such a mixed agent-swapping algorithm should always be preferred whenever the optimal composition of individuals in a multiagent system is unknown.