Adaption and Learning in Multi-Agent Systems