Robustness of multi-agent models: the example of collaboration between turmites with synchronous and asynchronous updating

Abstract : The robusntess of multi-agent systems to simulation conditions is analysed through a precise example, invented by Langton to investigate the foundations of artificial life. This system is composed of simple and memoryless agents, the turmites, which obey simple discrete local rules. While the local rules that govern each agent are kept constant, the interaction between agents are modified through nine variations. Our method consists in varying the updating scheme (synchronous vs. asynchronous) and the local conflict resolution policy (strong or weak exclusion rules). We experimentally estimate the effect of these modifications on three collaborative phenomena. We also analyse how the dynamics at the microscopic scale reflects the robustness of the system at the macroscopic scale. Observations confirm that the definition of the agents' behaviour is not the only setting that matters in the emergence of collaborative phenomena in complex systems: the way the agents are updated is also a key choice.